Skip to main content

Full text of "Human races"

See other formats

073 , 7 


Digitized by the Internet Archive 
in 2013 


Frontispiece. Urinary chromatogram of a 33-year-old Chinese woman. 

The arrow points to the area occupied by BAIB (/? amino isobutyric acid) 

an amino acid frequently excreted by individuals of Asiatic origin. 

For details see Figure 8 and Chapter III. 


Chairman, Physical Growth Department 

Fels Research Institute 

Professor of Anthropology, Antioch College 

Yellow Springs, Ohio 


Springfield • Illinois • U.S.A. 


Bannerstone House 

301-327 East Lawrence Avenue, Springfield, Illinois, U.S.A. 

This book is protected by copyright. No 
part of it may be reproduced in any manner 
without written permission from the publisher. 

Library of Congress Catalog Card Number: 60-14744 

With THOMAS BOOKS careful attention is given to all details of 
manufacturing and design. It is the Publisher's desire to present books 
that are satisfactory as to their physical qualities and artistic possibilities 
and appropriate for their particular use. THOMAS BOOKS will be true 
to those laws of quality that assure a good name and good will. 

Printed in the United States of America 


Aen years ago Carleton Coon, Joseph Birdsell and the 
present author collaborated on a little book entitled Races: A 
Study of the Problems of Race Formation in Man. In it we 
eschewed the then traditional anthropometric measurements and 
morphological ratings, and compiled no formidable catalog of 
human racial groupings. We were interested in one central prob- 
lem — how human races came to be. 

When we wrote Races, the mere mention of race was still un- 
comfortable to many, that soon after the tragic excesses of the 
Third Reich. But we were not concerned with notions of racial 
superiority or inferiority. We were writing about races in man, 
how they arose and how they changed, as they are changing still. 

Races was venturesome for its time, a time when the concept 
of a "pure race" was still tenable, and when scholars still wrote of 
fixed, static and unchanging races, incapable of genetic change. 
But the tempo of discovery soon passed us by. Critically investi- 
gated, using the new tools of biochemical genetics, human races 
proved capable of more rapid change than the most optimistic 
guess would have warranted. Directions of natural selection 
within race populations, a subject we had speculated about, 
proved most varied, and at the same time susceptible to exact 
measurement. With renewed interest in human raciation, prob- 
lems of human differentiation have been newly tackled. The en- 
tire field of Geographical Medicine, a newcomer among the dis- 
ciplines, has added vital meaning to the study of race. 

Human Races now is a very different book from what Races 
(1950) was. It is one mans product, both Coon and Birdsell 
being busy with their own investigations and their own publica- 
tions on race. At the same time, Human Races is the contribution 
of many investigators, the results of a most active decade of race- 


research. Some of these investigations have been reprinted in 
Readings on Race (1960), a companion source-volume to the 
present work, and therefore not recapitulated in detail here. 

Human Races is an attempt to describe what race is, and the 
mechanisms of racial differentiation in man. It will, I hope, help 
to dispel the antiquated notions of three "original" races, of the 
persistence of racial types, and of the role of undirected chance 
in bringing about racial differences. In their stead, I trust will 
emerge the contemporary picture of mans genetic response to 
local selective factors, the constantly changing nature of the 
natural populations we call races. 

While a more complete listing of indebtedness is given later in 
this book, I would like to thank Lois Conklin (who has lived 
through three books with me), Laura Newell, who has drawn 
illustrations, located references and corrected errors, and Dr. 
Lester W. Sontag and the Fels Fund for both tangible and in- 
tangible support. 

Stanley M. Garn 


Xhe author is indebted to Dr. Marvin Armstrong, Dr. 
Meinhard Robinow, and Dr. Henry Tomizawa for many helpful 
suggestions in preparing the present book. He is appreciative of 
permission of Dr. Anthony Allison, Dr. Baruch S. Blumberg and 
Dr. D. R. Roberts to reproduce several of their illustrations. The 
staff of the Antioch College Library, and particularly the librarian 
of the Fels Research Institute, were most helpful in completing 
the lists of suggested references and making photostats. Other 
illustrations not specifically drawn for Human Races are acknowl- 
edged as they occur. 

S. M. G. 



Preface v 


I. The Nature of Race 3 

The Contemporary Approach to Race 6 

Other Names for Race 8 

The Study of Race 9 

Suggested Readings 10 

II. Geographical, Local and Micro-races 12 

Race, "Race" and Race 12 

Geographical Races 14 

Local Races 16 

Micro-races 18 

Taxonomy and Research on Race 20 

Summary 21 

Suggested Readings 22 

III. Race Differences 23 

Pigmentation and Race 25 

The Hair 26 

The Bones .". 27 

The Dentition 28 

Growth and Race 29 

Physiological and Biochemical Differences 31 

The Haptoglobins 32 

Taste-blindness and Race 33 

Race Differences 34 

Summary 37 

Suggested Readings 37 



IV. Blood Groups and Race 39 

The ABO System 39 

The MNS-U System 41 

Rhesus and Race 43 

Duffy-an Australasian Gene 44 

Diego, an "Asiatic" Blood Group 45 

Blood Groups and Natural Selection 47 

Blood Groups and Human Taxonomy 49 

Summary 51 

Suggested Readings 51 

V. Natural Selection and Race 53 

Environmental Differences 54 

Pigmentation and Natural Selection 55 

Body Size and Natural Selection 56 

Body-build and Natural Selection 59 

Adaptations to Extreme Cold 62 

Adaptations to Night Cold 63 

Adaptations to Humid Heat 64 

Adaptations to Desert Living 66 

Summary 68 

Suggested Readings 68 

VI. Abnormal Hemoglobins, Malaria and Race 70 

Thalassemia and the Mediterranean 71 

The Genetics of Thalassemia 72 

Thalassemia and Malaria 72 

Sickle-cell Disease 73 

Culture, Malaria and the Sickle-cell Trait 77 

Culture, Malaria and Thalassemia 78 

Summary 79 

Suggested Readings 79 


VII. Race and Disease 81 

Kuru: Natural Selection and Sorcery 83 

Familial Mediterranean Fever 86 

Primaquine Drug Sensitivity 88 

Favism: When Gene Meets Bean 90 

Other Diseases and Race 91 

The Adaptive Nature of Hereditary Diseases 92 

Summary 92 

Suggested Readings 93 

VIII. Race and Genetic Drift 94 

Summary 98 

Suggested Readings 98 

IX. Race Mixture 100 

The Genetics of Race Mixture 101 

Race Mixture Viewed as Harmful 103 

Race Mixture and Hybrid Vigor 104 

Studies on Race Mixture 106 

Hybridization, Advantages and Disadvantages 107 

Summary 108 

Suggested Readings 109 

X. Race, Behavior and Intelligence 110 

Race and Temperament 110 

Race and Intelligence 112 

Summary 115 

Suggested Readings 115 

XL A Taxonomy for Man 116 

A List of Geographical Races 116 

A Selected List of Local Races 127 

Glossary — Index 133 


Figure Page 

Frontispiece — Urinary chromatogram ii 

1. The concept of race as type 5 

2. The concept of race as population 6 

3. A geographical race 14 

4. A local race 16 

5. Micro-races 19 

6. Percent of cases with Carabelli's Cusp 28 

7. Three areas of racial differentiation 30 

8. Diagram of a urinary chromatogram 32 

9. The three haptoglobin phenotypes 33 

10. Percentage of non-tasters in different geographical areas 35 

11. Percentage of Diego (Di a ) positives 46 

12. Relationship between mean body weight and mean 

annual temperature 58 

13A & B. Extreme differences in body proportions 60-61 

14. Adaptation to moderate cold during sleep 65 

15. The mechanism of "sickling" 75 

16. The frequency of the sickle-cell trait in East Africa 76 

17. The Fore territory in eastern New Guinea where Kuru 

is common 85 

18. The mechanism of genetic drift 95 

19. Race mixture 102 

20. Mathematical analysis of race mixture 107 

21. Polar-projection map showing geographical races 118 

22. Melanesians of New Guinea 120 

23. Spiral-tuft form of the body hair in an American 

Colored individual 123 


24. Suture bones on the skull of a New York State Indian .... 125 

25. Bushmen, show tight spiral-tuft hair 126 

26. World map showing location of the thirty-two selected 

local races 128 






early three hundred years ago, Carolus von Lin- 
naeus, the great naturalist and taxonomist, set up his famous 
classification of living things. When he came to man, Linnaeus 
properly assigned man to the order Pwnates on the basis of 
numerous and fundamental biological similarities. To the genus 
that contained man, he gave the traditional Latin name Homo. 
And, having weighed the evidence for and against several species 
of man, von Linnaeus assigned all living forms of mankind to one 
species within the genus Homo, as Homo sapiens. 

Today, we know far more about man than Linnaeus did. We 
have recovered from the Pleistocene deposits of Java and China 
fossil species of Homo that are quite distinct from Homo sapiens. 
We have come to study many groups of living men quite unknown 
in Linnaeus' time. Beyond the simple descriptions available to 
the Swedish taxonomist, we have precise anthropometric measure- 
ments, data on blood groups, the haptoglobins and many measures 
of biochemical functioning. Although there are some traits in 
which different human groupings show little overlapping, living 
mankind clearly constitutes a single polytypic species. Fossil non- 
sapiens hominids no longer exist, and we are all Homo sapiens, 
as assigned by Linnaeus. 

But, within this single species which now covers the habitable 
globe there are many discrete groupings, some so clear-cut as to 
be obvious to the least- trained observer, and others less easily 
distinguishable except after intensive study. These groupings, 
differing greatly in size and taxonomic status have commonly been 
lumped under the single term "race." Thus, some so-called races 
are grossly distinct by all of the tests we now have and use, while 
other groups called races differ in smaller degree, in the averages 


of certain measurements, in the proportions of discrete traits, and 
in the frequencies of such biochemical differences as the several 
blood groups. 

In addition to races defined by zoologists, anthropologists and 
human geneticists, human groupings of various kinds have been 
dignified or designated by the term "race." Race has been equated 
with language, and that is the sole useful meaning of the "Aryan" 
race. Race has been identified with religion, as in the case of the 
Jewish "race," which in reality comprises a number of discrete 
populations, some quite unrelated to the others. National group- 
ings have frequently been called races, especially in periods of 
intensive nationalism. While at times linguistic groupings and 
biological races may coincide and while religions or even national 
boundaries may delimit race-populations of various sizes, lan- 
guage, religion and national affinity are hardly measures of race. 
Race is a biological concept and races are biological units. Races, 
moreover, are natural units and not artificial assemblages created 
by selecting "types" out of a population. 

A century ago many scholars believed that all human groups 
could be explained in terms of a few "original" races, usually 
numbering three. Many attempts were made to explain known 
human populations as mixtures, in varying degree, of these hypo- 
thetical original races. In the days when all dark-skinned peoples 
were accepted as "Negroes," when all straight-haired people were 
accepted as members of the "yellow" race, an original set of but 
three races satisfied the data and fitted the assumptions. Yet, it is 
now obvious that such a working hypothesis is untenable. No 
possible combination of Negro, Mongoloid and White could pro- 
duce the Australian aboriginal. No such combination could ex- 
plain the American Indian. The blood group distributions in 
Melanesia, Micronesia and Polynesia could hardly be explained 
in terms of Negro-Mongoloid- White admixture. In fact, there is 
little reason to believe in a system of but three original races, 
that but three original races ever existed. 

In Europe too, scholars formerly postulated the one-time ex- 
istence of a limited number of "pure" races, which through ad- 
mixture, gave rise to the complex situation we see from Finland 
to North Africa today. As proof of their hypothesis, they pointed 


to the occurrence of individuals exemplifying the characteristics 
attributed to the hypothetical pure races. From blue-eyed, long- 
headed, light-haired men and women they inferred the "Nordic" 
race, and from various combinations of features, found in indi- 
viduals, other ancestral "pure" races were similarly deduced. 

Yet, while the individuals themselves, as "types" clearly exist, 
the inference that they recapitulate ancestral strains can be chal- 
lenged. As Edward E. Hunt, Jr. (1959) has demonstrated, the 
individual "types" are merely chance combinations of the geneti- 
cally independent traits. Since the traits are independent, and 
"segregate" out separately, such types as may be found in a popu- 
lation prove nothing about the appearance of ancestral groups. 
Blue eyes and blond hair are no more proof for an original Nordic 
race than red hair and freckles point to an original Rufous race, 
or short stature and heavy beards to a race of Trolls. 

In its heyday, however, the approach to race as a series of 
types, led to racial typing of indivduals, with rather remarkable 
results. Various members of a single family were often assigned 
to different "races," and three brothers or three sisters could be 
typed as belonging to as many different "races" ( Fig. 1 ) . 

i Frederic Jones 1 

Fig. 1. The concept of race as type. In this approach, where individuals 
are classified as to appearances, brothers may be assigned to different "races." 




The contemporary approach to race in man, as in other living 
forms, is eminently simple. It is a return to the basic principles 
of classification, to the days before hypothetical original races 
were postulated, and populations were analyzed in terms of vary- 
ing proportions of different racial "types." Yet the contemporary 
approach to race stems from population genetics, where a race is 
viewed as a breeding population, neither more nor less. A race in 
man, as in any living form, is a population, sl population of men, 
women and children, of fathers, mothers and grandparents (Fig. 
2). Members of such a breeding population share a common 
history, and a common locale. They have been exposed to com- 
mon dangers, and they are the product of a common environment. 
For these reasons, and especially with advancing time, members 
of a race have a common genetic heritage. 

2. The concept of race as population. In this approach, emphasis is on 
the biological race or population isolate. 


Such a definition, race as identical with the breeding popula- 
tion, transcends history. One race, or one breeding population 
may go back 20,000 years, or even more. The Central Austra- 
lians, and possibly the Andamanese, can claim such antiquity, 
with little or no admixture. Another race, as for example the 
American-Colored, may have been formed by admixture in recent 
memory, but it has equal claim to racial status. A third race, far 
smaller than either of the two mentioned, but no more recent than 
the American Colored, is the genetically and geographically iso- 
lated population of Pitcairn Island. If there is hesitation at calling 
the Pitcairners a race, how large must a population be to be a 
race? If admixture, both known and recent, provides a stumbling 
block to acceptance, how about the probable incorporation of 
Neanderthals into the European population, and the certain addi- 
tion of late-Paleolithic survivors as late as 8,000 years ago? 

From the standpoint of taxonomy, that is classification, how a 
race was formed is irrelevant. A race is a race is a race whether it 
goes back unchanged for six millennia, or whether it resulted from 
admixture after 1850. Our preoccupation with recent history may 
make us view the Neo-Hawaiian race with hesitation, but how 
about races whose mixed origins barely antedate the written 
record? In similar fashion, there is no number test for race. The 
Bushmen of South Africa, totalling perhaps 25,000, are quite as 
real a breeding population as the American Colored, who number 
some 15 million. 

Moreover, there is increasing evidence that races do change in 
their genetic makeup, even in the absence of admixture. The 
sickle-cell gene, for example, has been on the increase in Africa 
ever since slash-and-burn agriculture was introduced. With drain- 
age, mosquito control, DDT and antimalarial drugs, the sickle-cell 
gene will decrease in generations to come. Differences between 
races once thought to go back to remotest antiquity, now appear 
to be genetic responses to environmental conditions. Recognizing 
that changes do occur within races, that races are not fixed in 
their genetic makeup, our attention has shifted from the simple 
existence of races, and the description of the differences, to the 
mechanisms that bring these changes about. 



So far in this book the term race has been used to designate 
natural human populations, the sum of which constitutes the 
species as a whole. To be technical, the "race" is the taxonomic 
(that is classificatory ) unit immediately below the species. Thus 
it is that the single species Homo sapiens comprises a number of 
races, of varying magnitude, as will be discussed in the next 
chapter. The majority of physical anthropologists, taxonomists 
and geneticists agree with this usage, holding to the term race, 
despite misuses that have occurred in the past. 

A minority of workers, however, prefer other terms for the 
same taxonomic units. The word race, they argue, has been ill- 
used by politicians and demagogues, by Madison Grant, Lothrop 
Stoddard, de Gobineau, and by Adolph Hitler. Some "races," as 
Sherwood L. Washburn has pointed out, never existed. Like the 
"Nordic" race of Germanic fame, they were hypothetical formu- 
lations that had no bearing in fact. It would be advisable, accord- 
ing to one argument, to use a new term for race, free of unfortu- 
nate historical associations, and which would mean exactly what 
we want it to mean. 

Various euphemisms for race have been suggested, among them 
"variety," and "division." The geneticist Hans Kalmus (1950) 
prefers the word "strain," a term used by animal breeders and 
plant geneticists. Ashley Montagu ( 1951 ) in much of his recent 
writings, has championed the term "ethnic group" as a substitute 
for the familiar word "race." 

However, these substitute terms have defects of their own. A 
botanical "variety," such as the dwarf Forsythia, is not equivalent 
to a polymorphic human population. A strain or breed is an arti- 
ficial line, for the most part maintained by selective breeding. 
The term "ethnic group," useful in its original sense of a cultural 
group, creates confusion when equated with biological races, the 
more so when a list of "ethnic groups" proves virtually identical 
with other lists of races! 

To be sure, it is not tremendously important what term we use 
for race, and here we may recall Shakespeare's comment on the 
rose. A race-population remains a race-population whether called 
breed, variety, strain, or ecar, which is simply race spelled back- 


wards. A newly-minted, specially-coined word might be free of 
unfortunate associations yet would contribute little to the ameli- 
oration of racial prejudice. Under the circumstances we might 
as well continue to talk about races, devoting our attention to 
spelling out the exact taxonomic category under consideration. 


The real question, of course, is why we study race, why we are 
interested in raciation in man, and why we expend valuable time 
and considerable money in the investigation of such seemingly 
esoteric subjects as the blood groups of the Idaho Basques. 

For some the existence of races is a challenge, as with moun- 
tains that beg to be climbed simply because they are there. In 
much the same vein, an isolated race, unsullied and unstudied, 
still beckons with romantic appeal. As with Richard Burton, there 
are always some who will brave dangers and risk their own for- 
tunes to be the first to visit and describe a previously unstudied 

Still others are interested in taxonomy, that is classification. 
And to have a complete taxonomy, that is a complete classifica- 
tion, all human populations must be seen and studied. We are far 
from a complete taxonomy for man: part of Africa, much of Asia 
and a surprising portion of South America is still to be investi- 
gated with care. In the meanwhile, new race-populations are 
springing up, and these too merit analysis. 

Another reason for studying race is phylogenetic. How are 
races related to each other, and which races arose from others? 
In some cases we have historical data to guide us as with the 
Cape Colored, and the American Colored, the Ladino and the 
neo-Hawaiian. Even so, serological data are necessary in order 
to complete our knowledge as to proportions entering into ad- 
mixture. In other cases we own mere educated guesses. Clearly 
the American Indians stem from Asiatic Mongoloids but when 
and where? Are the Bushmen ancient, going back to the earliest 
origins of Africa, or are they merely desert-adapted Negroes? Con- 
versely, are the Zulu and Bantu themselves evolved Bushmen? 

These latter problems bring us to the more intriguing reasons 
for investigating race, including the question of how races and 


particular races came to be. Why are Negroes dark, and are there 
similar explanations for the peoples of Southern India, and the 
Melanesian Islands? Why do Asiatics have straight, coarse black 
hair and sparse beards? Why are the Papago fatter than the 
Navaho in the same region? Why are the Basques so peculiar in 
their Rh blood group distribution? Why are the Pygmies pygmy, 
the Blackfoot Indians tall, and why are the Aleut short? 

Why are certain genetic diseases limited to some groups and 
absent in others? Since such diseases are deleterious and subject 
to natural selection, what explains the continuance of sickle-cell 
disease in Africa, Favism and Mediterranean anemia in Southern 
Europe, or Kuru in New Guinea? And turning from disease to 
physiology, are the Eskimo cold-adapted and are the natives of 
Dakar or Brazzaville adapted to both heat and high humidity as 
would be reasonable to suppose? 

Such questions as these are no longer excursions into the un- 
answerable. Within the past decade we have seen some of them 
explained, and others close to understanding. How races came to 
be is no longer a philosopher's conundrum, and the peculiarities 
of particular races come closer and closer to comprehension. The 
distinctive characteristics of every race may now be understood 
in terms of the special environments in which they have lived. 

The study of race bears a personal attractiveness to us. This is 
our species, and as men we are inevitably fascinated by man. 
Now, quite suddenly we are in a position, as many investigative 
fields come to maturity, to answer the fundamental questions that 
will lead to a more complete understanding of the different races 
of mankind. 


Anonymous: The Race Question in Modem Science, UNESCO publi- 
cations, New York, Morrow, 1956. 

Count, E. W.: This Is Race, New York, Shuman, 1950. 

Dunn, L. C: Heredity and Evolution in Human Populations, Cam- 
bridge, Harvard University Press, 1959. 

*Garn, S. M. (ed.): Readings on Race, Springfield, Thomas, 1959. 

Hunt, E. E., Jr.: Anthropometry, genetics and racial history, Am. 
Anthropol, 61: 64-87, 1959. 


Kalmus, H.: Genetics, Harmondsworth, England, Penguin Books, 1950. 

Montagu, M. F. A. : An Introduction to Physical Anthropology, Spring- 
field, Thomas, 1951, 1960. 

Thieme, F.: The population as a unit of study, Am. Anthropol., 54: 
504-509, 1952. 

* Suggested readings contained in Readings on Race are designated 
by an asterisk throughout this book. 



Ahe original edition of Races published in 1950, con- 
tained a listing of thirty living race-populations in man. In the 
very same year, in his Genetics and the Races of Man, William 
C. Boyd described a total of six races: however, one of them was 
both hypothetical and extinct. Though Boyd (1958) has more 
recently augmented his list of races, raising the number to thir- 
teen, we are obviously dealing with different orders of magnitude. 

Other listings of human races have run the numerical gamut. 
A century ago some workers divided mankind into but two races, 
a straight-haired race and a woolly-haired race. Most physical 
anthropologists have described from twenty to fifty distinct races, 
though some of these races refer to individual "types" and not to 
breeding populations. Specialists, working with particular areas 
of the world have been even more generous in their race-assign- 
ments, granting fifty or more for Europe alone ( cf . Coon, 1950 ) . 

Faced with such a perplexing situation, such a range of num- 
bers of races, it is customary to refer to "lumpers" and "splitters." 
Lumpers, among taxonomists, are those who group a number of 
distinct varieties into one broader, larger category, explaining that 
the differences are too trivial to warrant so complex a taxonomy. 
Splitters, on the other hand, take the microscopic rather than the 
macroscopic view holding that any distinct variety merits atten- 

But the situation in regard to man is not simply a matter of 
lumping or splitting: rather it is due to an overly elastic use of the 
term race. For some workers, such as Boyd (1950) races are 
identical to geographically-delimited collections of races. Prac- 
tically, Boyd's "races" are identical with the "stocks," "divisions," 



"primary races" or "major races" as defined by previous workers. 
To other human taxonomists, however, particularly those influ- 
enced by population genetics, a race is a population. Inevitably, 
therefore, equating "race" with population-collections results in a 
smaller number of races, while restricting the term race to actual 
populations results in a far larger number of races. It is as if the 
term regiment were variously employed to refer to platoons, 
squadrons, brigades and armies. 

One expedient would be to coin a new set of words for taxo- 
nomic units smaller than the species. One might have the species, 
then the stock, then the breed, then the variety, etc. But there 
are dangers in seting up a completely new terminology, as men- 
tioned in the last chapter. The most practical suggestion is one 
made by Bernhold Rensch, the German systematist, in 1929. He 
uses the term geographical race, to describe the broad, geograph- 
ically-delimited population collections, and the term local race 
to refer to race-populations themselves. As Ernst Mayr (1950) 
puts it "this system facilitates communication without encumber- 
ing nomenclature." 

While geographical races and local races adequately distinguish 
Bushmen, for example, from Africans in general, or Navahos or 
Hopi Indians from the broad category of Amerindians, these two 
terms are not quite enough to fit all of the data. The population 
of Oslo is genetically distinct from the population of Helsinki, yet 
neither is a true breeding population, a genetic isolate. Salerno 
and Padua are distinct in many respects, yet there is no fence 
about Salerno, and no moat surrounding Padua. Here Dobzhan- 
sky's term microgeographical race comes to our assistance in de- 
lineating statistically-distinct populations which cannot be de- 
limited as circumscribed breeding-populations. However, and 
with apologies to Professor Dobzhansky, the term micro-race will 
be used throughout this book (instead of micro-geographical 
race) simply because it is less confusing, less likely to be con- 
founded with geographical race. 

Geographical races, local races and micro-races, these do not 
encumber nomenclature. Whosoever uses them is immediately 
aware of the fact that he is dealing with race, the taxonomic unit 
immediately below the species. These terms facilitate commu- 



nication, in that they explicitly state which taxonomic unit is in- 
volved. With respect to geographical races, there is no plethora 
of them. As race-collections they certainly do not exceed ten. Of 
local races there is obviously a multiude numbering surely into the 
hundreds. The Navaho, Hopi, Zuni, Pima, Papago, Cocopa, 
Haida, Salish . . . these are a few of the local races among the 
American Indians. And micro-races, in densely populated areas 
of Europe and Asia — these run into the thousands, each hamlet 
being genetically somewhat distinct from the others. 

Clearly, the one term race is not enough for us to use. By being 
more explicit we gain clarity and lose confusion. 


The geographical race is the largest of the three categories of 
races and encompasses (in each geographical region) the other 
two. For a geographical race is by definition, a geographically- 
delimited collection of similar races (Fig. 3). To a large extent 
the geographical races of mankind coincide with the major con- 

Fig. 3. A geographical race— a collection of race populations, separated from 
other such collections by major geographical barriers. 


tinents, and are therefore identical with continental races, as the 
term is used by Boyd and others. However, geographical races 
may also be spread over major island chains, as is evident in the 
Pacific today. 

The existence of geographical races is due, of course, to the 
great geographical barriers, chief among them oceans, that form- 
erly limited the expansion and migration of local races and pro- 
tected them from the introduction of different genes. Thus, in 
pre-Colonial South America, there was little or no gene-introduc- 
tion from either Africa or the Pacific. Gene flow in and out of 
South America was funneled through the narrow isthmus of Cen- 
tral America. The great sub-continent of Australia also represents 
a situation where geographical race and geography coincide, due 
to water barriers all around, and no major tradition of navigation 
and sea travel. 

However, the continents marked out in different colors on the 
map do not perfectly delimit geographical races, whereas the 
geographical barriers to human migration do. Africa is separated 
from Europe, and its own northern region, by great ranges of 
desert, scarcely inhabited by a few wandering tribes, and by the 
Atlas mountains. South of the Sahara and through Africa to its 
southernmost tip there is one geographical race, comprising a very 
large number of local races, whereas North Africa is racially con- 
fluent with the Near East and Europe. 

Similarly, the eastern limit to the geographical race inhabiting 
Europe is in Western Asia, in the scarcely inhabited uplands, 
and not coincident with conventional continental divisions. How- 
ever, the high and uninviting mountains that mark the Tibeto- 
Indian border on the maps have long restricted population inter- 
change to a thin trickle. The facts of geography, the mountain 
ranges, the deserts and the oceans have made geographical races 
by fencing them in. 

Within each geographical race the individual populations re- 
semble each other more or less. In the aggregate, resemblances 
within geographical races are far greater than those between 
them. However, intra-population differences are still great, espe- 
cially taken trait by trait or gene by gene. In each geographical 
race there are tall populations and short populations, heavy-set 



groups and linear groups. Certain human differences transcend 
geographical race, and are more meaningfully distributed with 
respect to climate or disease ( Chapter VII ) . 

A geographical race is a collection of populations whose simi- 
larities are due to long-continued confinement within set geo- 
graphical limits. 


In contrast to geographical races which are geographically 
delimited population collections, local races correspond more 
nearly to the breeding populations themselves. Whether isolated 
by distance, by geographical barriers or by social prohibitions, 
local races are totally or largely endogamous, and the very small 
amount of gene-flow ordinarily comes from contiguous and re- 
lated local races ( Fig. 4. ) . 

The Bushmen of South Africa are one example of a local race 
where the territorial limits are defined, and where breeding has 

Fig. 4. A local race— a breeding population adapted to local selection pres- 
sures and maintained by either natural or social barriers to gene interchange. 


been confined almost exclusively to the local race itself. The sev- 
eral native local races of Australia also typify the situation, as do 
the Ituri-forest Pygmies. Though the latter have contributed 
wives to the taller Negroes around them, gene-flow appears to be 
largely one-way, and the Ituri-forest Pygmies constitute a true 
breeding population to the present day. 

Clear-cut local races such as these are largely independent 
evolutionary units, and as such are of particular interest. Other 
examples of local races include the Yemenite Jews, isolated re- 
productively from their Arab neighbors, and from other Jewish 
populations for millennia. Whereas the Yemenite Jews, now 
being absorbed into the Europeanized population of Israel, have 
religious affinities with the European, North African, Kurdistani 
and Oriental Jews, their status as a separate local race held for 
thousands of years. 

As a further example of local races, one may consider the vari- 
ous Eskimo populations widely spread from Greenland across the 
Arctic to Alaska, the Aleutians and Siberia. Each has been sepa- 
rated from the other for millennia. It is questionable whether one 
Greenland Eskimo got to Alaska in the last five-hundred years. 
As to the Aleut, despite their proximity to Alaska, well under 1% 
of Alaskan genes have found their recent way into Umnak, 
Atka, or the Pribilofs. 

Local races are most easily identified where populations are 
relatively small, and there is little doubt as to their limits as indi- 
cated by geographical separation or by cultural prohibitions on 
marriage outside of the group, as with the several Gypsy popu- 
lations of Europe. Local races can also be delineated, though 
less neatly in the populous areas of the world. By way of ex- 
ample, the demographic populations of Northwestern Europe and 
Southern Europe share markedly different histories and are, on 
the whole, quite distinct. The former population, relatively late to 
expand, has done so following the discovery of the New World 
and the subcontinents of the Pacific and has poured into these 
territories. Thus it is that North America, New Zealand and 
Australia constitute territorial extensions of Northwestern Europe 
from a racial point of view, while Central and South America is 


more of an extension of the breeding population of Southern 

The Northwest Europeans, though constituting a smaller taxo- 
nomic unit than the European Geographical race, are not as neat 
a population as those mentioned earlier in this section. Numerous 
special problems interpose themselves, the problem of differential 
migration (who migrated?) and differential selection in the new 
and at least temporarily hostile environments. Moreover, even 
among the stay-at-homes, there are local differences, as shall be 
mentioned shortly under micro-races. 

Nevertheless, it is the local race that we view and measure 
somewhat more easily when the numbers are small, whereas the 
geographical race represents more of an abstraction. The popula- 
tion as a unit-of-study is identical with the local race, and becomes 
increasingly more difficult to investigate as it becomes less easy 
to delineate. 


For much of the world today, as in Europe and Eastern Asia, 
neat local races are hard to come by. Except for a few populations 
in Europe such as the Basques, or the Lapps, one cannot define a 
local race by an ethnographic survey. A man from Berlin marries 
a woman from Stuttgart, her brother lives in Hamburg and has 
married a Dane from Copenhagen, whose sister now resides in 
the Finnish University city of Turku (but is on leave in Cleve- 

Nevertheless, there are very real differences in the genetic 
makeup of cities, and continual changes in the frequencies of 
various genes in either the north-south, or the east-west direction. 
Some of the differences are so apparent that we can divide Europe 
into a series of local races, Northwest European, East Baltic, 
Alpine, Mediterranean, etc. Other differences are more subtle, 
visible to the trained observer, or detectable by the serologist or 
biochemist from his data. 

Regional differences in Europe are in part due to ancient 
settlement patterns, and to the local perservance of local races. 
Even though the genetic insularity of the old city-states has long 
been breached, propinquity is still a very real determinant of 
mating. "International" marriages, especially in the academic and 



professional classes, should not blind us to the fact that marriage, 
or mating, is a mathematical function of distance. With millions 
of potential mates, the male ordinarily choses one near at hand. 
In fact, there is a third kind of genetic isolation. In addition to 
geographical isolation, as between continents, and cultural isola- 
tion, as between local races, there is isolation by numbers. The 
denser the population, the more nearly the boy marries the girl 
next door ( Fig. 5 ) . 

This latter phenomenon, which somewhat deflates the picture 
of the romantic human male, and allies him more nearly with the 
field mouse or mosquito (which have similar mating ranges) has 
the effect of maintaining micro-races. It will be eons, at the 
present rate, before Copenhagen is no longer different, genetically, 

Fig. 5. Micro-races. In a densely populated area local races may not be 

demonstrable, yet biological distance may maintain regional differences as 

in the centers A, B and C shown here. 


from Oslo or Stockholm, or Venice from Naples or Rome. Further- 
more, local selective factors will continue to be at work. Thus, 
one genotype will be favored in this city and another in that, 
maintaining and even exaggerating the genetical differences that 
now exist between micro-races. 


Given geographical races, local races and micro-races, there 
may appear to be some question as to their relative importance. 
From one point of view, geographical races may seem to be of 
greatest interest. Geographical races are large, and there are so 
few of them. From another point of view local races may be 
favored. After all, local races are natural populations (not col- 
lections of convenience); they are the basic evolutionary units 
and they can be studied in divers ways. 

Actually, the importance of these successive taxonomic cate- 
gories depends very much on the problem at hand. By way of 
example, the ancestry of the American Indians, and their relation- 
ship to Asiatics focuses attention on two geographical races. 
Differences between Europeans and Asiatics with respect to blood 
group B, the Diego factor, or the Rhesus-negative gene, again 
involve geographical races. The Polynesians, and possible ex- 
planations for their polymorphism, necessitate attention to con- 
tiguous geographical races, in Asia, in Melanesia and Papua, and 
in Australia. 

At the same time, the diversity of populations within each geo- 
graphical race introduces problems of its own. If we are to com- 
pare Amerindians to Asiatics, with an interest in common origins, 
which Amerindians and what Asiatics shall we compare? Shall 
we used weighted average values, which may be biased by par- 
ticular populations, or try to select prototypical Amerindians and 
Asiatics in which case our comparisons obviously reflect the 
populations selected for use? 

In contrast to geographical races, local races are both easier to 
define and simpler to investigate. Within the course of a year 
it is possible to measure and blood-type all living Aleut. It is 
possible to obtain a fair sample of all known Bushman bands. 
Moreover, with local races, we are interested in local selective 


factors; diet, disease and environmental stresses. Local races, 
therefore, commend themselves to the investigation of evolution- 
ary forces. The degree of out-marriage can be determined in 
assaying the role of "drift." Survivors can be studied to determine 
possible directions of selection. With local races inhabiting a fixed 
territory, and where their neighbors can be examined as well, the 
role of admixture may be quantified. Local races, therefore, offer 
the maximum opportunity for evolutionary studies. 

With micro-races, our human material approaches more nearly 
the demographic population rather than the natural or race- 
population. Micro-races are not delimited by geographical or 
even tribal barriers to gene flow. Nevertheless micro-races offer 
numerous opportunities to investigate the mechanisms of differ- 
entiation. In Saudi-Arabia, as in the more populous parts of 
Africa, the incidence of malarial infestation may be related to 
local differences in the frequency of the sickle-cell gene. In 
European cities it is possible to investigate differential survival 
in the face of smog, a problem we have begun to consider in 
Donora, Pennsylvania, Los Angeles and other American com- 
munities. Differential mortality and morbidity brings about 
genetic changes within populations, and micro-races offer the best 
opportunities at present for such comparative studies. 

With respect to geographical races, local races and micro-races 
the question is not which is more important, but rather what 
questions we are trying to answer. 


The number of races of mankind which varies from no more 
than two to several hundred according to the taxonomy consulted, 
ceases to pose a major problem if the taxonomic category used 
is precisely defined. 

Immediately below the species is the geographical race, a 
geographically- delimited collection of local races which may 
differ markedly, one from the other. Local races, in turn, corre- 
spond to natural or breeding populations, and are at once the 
units of evolutionary change and the common subjects for in- 
vestigation. Micro-races, though not isolated geographically or 


by extensive cultural prohibitions, still differ from each other in 
numerous ways. 

Geographical races, local races and micro-races offer opportu- 
nities for very different investigations in relation to race. One is 
not more real or more fundamental than the other, but each pro- 
vides the answer to different questions and the solution to differ- 
ent problems of ongoing evolution in man. 


Boyd, W. C: Genetics and the Races of Man, Boston, Little, Brown 

and Company, 1950. 
*Boyd, W. C: Genetics and the Races of Man, Boston University 

Lecture, Boston, Boston University Press, 1958. 
Coon, C. S. : Races of Europe, New York, Macmillan, 1939. 
Coon, C. S., Garn, S. M. and Birdsell, J. B.: Races. A Study of the 

Problems of Race Formation in Man, Springfield, Thomas, 1950. 
Dobzhansky, Th.: Genetics and the Origin of Species, Chapter VI, 

New York, Columbia University Press, 1954. 
*Garn, S. M. and Coon, C. S.: On the number of races of mankind, 

Am. Anthropol, 57:996-1001, 1955. 
Mayr, E.: Taxonomic categories in fossil hominids, Cold Spring Harbor 

Symposia on Quantitative Biology, 15:109-118, Cold Spring Harbor, 

The Biological Laboratory, 1950. 
Rensch, B.: Das Prinzip geographischer Rassenkreise und das Problem 

der Artbildung, Berlin, Boentraeger, 1929. 
* Reprinted in Readings on Race. 



xXaces differ in a great many respects — in language, in 
dress, in gesture, in body size and proportions, in amino-acid 
excretion patterns, in tooth morphology and in the proportions 
(or frequencies) of the various blood groups. 

Those differences that are primarily learned, such as language 
or religion, may play an important role in maintaining the genetic 
isolates we call races. Languages are effective isolating mecha- 
nisms, as are religions. In various parts of the world there are 
sympatric races, occupying the same territory, but reproductively 
isolated because of religious practices or language differences. 
Yet, such learned differences in behavior are not biologically in- 
herited: they may be concommitants of race, but are not a part 
of race. 

Besides such learned differences between races, clearly non- 
genic in nature, there are many differences in body size and form 
that are responsive to environmental alterations. Stature is a type 
example. Though stature is in part inherited, nutritional ade- 
quacy during growth affects stature to a very large degree. Head- 
form, once considered a purely inherited trait has since been in- 
cluded among the "plastic" human features, susceptible to nutri- 
tional modification or to the effects of cradling and other skull- 
deforming practices. Such plastic traits, though showing consider- 
able differences from population to population, are of question- 
able utility in racial comparison except when great caution has 
been used in untangling the effects of nurture from genetic nature. 

The ideal traits for use in racial comparisons, and in the analysis 
of ongoing evolution within races, are those that are simply in- 
herited and of known mode of inheritance. With such traits it is 
possible to go beyond trait frequencies (the proportion of indi- 



viduals showing the traits in question) to gene frequencies (the 
proportion of the allelic genes in question ) . * Gene frequencies, 
though mathematically derived from trait frequencies make for 
more accurate comparisons and facilitate calculations. That is 
why the blood groups, the haptoglobins, taste-blindness and the 
abnormal hemoglobins, finger prints and rare hereditary diseases 
are so useful in the study of race: knowing the mode of inheri- 
tance, phenotype (or trait) frequencies can be supplanted by 
gene frequencies. 

Nevertheless, there are many racial differences, not simply in- 
herited, but due to the cumulative effect of several genes, that are 
of marked utility in population comparisons. Such "polygenic" 
traits include hair form and skin pigmentation (excepting albi- 
nism ) , tooth form, and the extremes of body build. Not knowing 
the mode of inheritance, gene frequencies cannot be calculated 
and the mathematics of population comparisons is therefore 

A further caution must be added about purely phenotypic 
differences, even if clearly inheritable. A given phenotype may 
be due to one set of genes in one population, and another set of 
genes in a second population. It is by no means certain that the 
genes for dark skin are the same in all populations and therefore 
the assumption of genetic affinity between the dark-skinned peo- 
ples of Africa and Melanesia is questionable. This caution applies 
with even greater force to such complex phenotypes as nose form 
or the extremes of stature. It is questionable whether the various 
"pygmies" of the world are related simply because they are 
pygmy, and it is a very tenuous assumption that the Ainu are re- 
lated to Europeans simply because of their generalized hirsutism. 

Thus it is that the neatly and simply inherited differences are 
currently of maximum use in the study of race, while other obvi- 
ously inheritable but polygenic traits are of secondary utility until 
more is known about their genetics, and the appropriate mathe- 
matical methods have been developed. 

* Individuals of blood type O are homozygous for the gene, being of the genotype 
OO. But the majority of individuals of blood type A or B are of the genetype OA 
or OB. The gene frequency for O is therefore greater than the trait frequency. In 
the case of Mendelian dominants the situation is reversed. For computational methods 
see Boyd (1950) appendix A. 



Nevertheless, racial differences in skin pigmentation deserve 
first mention, if only because such differences have been long 
observed. Among the lightest-skinned individuals of northern 
Europe there is little of the brown-black pigment called melanin 
in the lower layers of the skin, and the apparent color of the skin 
is largely due to reflections from the skin surface, with some of 
the blood pigments showing through. So-called "whites," of 
course are not white at all, but a light pinkish brown, reflecting 
from less than 20% of the light ( in the blue end of the spectrum ) 
to nearly 40% in the fairest. The "browns" and "blacks" have 
increasing amounts of melanin and in the darkest skins, no more 
than 1% of the light may be reflected from the unexposed skin 

To the "best of our knowledge, skin pigmentation is primarily 
dark melanin in various amounts and in various degrees of dis- 
persion. There are also yellow and yellow-red pheomelanins, but 
not known for man except perhaps in red hair follicles. At any 
rate there is no true yellow-skinned race and we need not invent 
a complicated ancestry for the Bushmen who have been described 
as having "yellowish" skins. 

The amount of pigment in the unexposed skin, of course, is 
only half of the story, for the capacity to tan is also important, 
and this capacity is not invariably related to the unexposed skin 
color. Very little has been done on this tanning capacity but it 
figures large in racial appearances. A well-tanned Norwegian or 
Swede may be darker than an indoor-working American Negro. 
The light skins of aristocratic Polynesians or Arabs, need not be 
indicative of genetic differences but merely differential exposure 
to sunlight. On the other hand, the capacity to tan may be limited 
in even brunette-skinned individuals. Therefore racial compari- 
sons must be limited to areas of skin ordinarily unexposed to the 
sun's rays or made in connection with controlled tanning studies. 



Next to the skin and its pigmentation, hair form, color and 
abundance have been most often utilized in racial taxonomies. 
In fact, hair form alone would effectively discriminate two-thirds 
of the world's populations. 

The range of hair forms is wide, and in this respect, man is 
more variable than any other primate, though in all fairness no 
other primate occupies so wide a geographical range! In much 
of Asia and in aboriginal America, the head hair is straight or 
nearly so and coarse ( over 100 microns ) . In much of Africa head 
hair is highly curved, even to the tight centimeter-wide spirals 
best seen in the Bushman and Hottentot. In Europe there is a 
wide range of hair forms, from nearly straight — but rarely of 
Mongoloid coarseness — to helical, as seen in Greek statues of 
Pericles' time. A world survey of head-hair form is given below. 


(By Geographical Areas) 

Straight Straight Helical to Spiral 
Geographical Area Coarse to Wavy "Woolly" Tuft 

North and South America 



Papua — New Guinea 





Hair on the body, less extensively studied, is characterized by 
great racial variability. In most of Asia, America and much of 
Africa, body hair is sparse or absent. In Europe, and the Middle 
East to Afghanistan and Pakistan, body hair is often well-de- 
veloped. Sporadically, generalized hirsutism exists — in Papua and 
New Guinea, among the Australian aboriginals, and in the Ainu 
of northern Japan. However, the hairiness of the Ainu has been 
exaggerated as a brief stay at any American bathing beach will 

Balding, to be more exact, male pattern balding constitutes a 
very real racial difference. Hormone-mediated and gene-de- 





















termined, it is rare among Asiatics, Amerindians and Africans, and 
common in Europe and the Middle East. Balding is associated 
with, but genetically independent of generalized hirsutism. At 
its extreme manifestation, male pattern balding begins in the 
early twenties. Possibly, if Italian data hold for other countries, 
early male pattern balding is associated with increased fertility. 


Racial differences in the size, proportions, form and mineral 
content of the bones are well documented for a number of groups, 
since these variations have been extensively studied on skeletons. 

Using trunk-length as a reference, the relative proportions of 
the limb bones vary markedly from Eskimo and Japanese, at one 
extreme, to American Negroes and especially certain African 
groups at the other extreme. Racial differences in the relative 
lengths of the metatarsal bones of the feet as expressed by differ- 
ing digital formulae are obvious even in the living, as are racial 
difference in the calcaneus and astragalus. 

The presence of accessory or "Wormian" bones along the suture 
lines of the skull, typically between the occipital and parietal 
bones where they meet at the back of the head, is characteristic 
of Asiatics and especially American Indians ( Fig. 24 ) . This alone 
is a great help to law-enforcement officers, when presented with 
an exhumed skeleton to identify. Accessory or suture bones dis- 
tinguish recent white burials from older Amerindian remains. 

Medically, Negro-white differences in the volume of the sacral 
canal become of importance when certain types of spinal anas- 
thesia are contemplated. Racial differences in the architecture of 
the pelvic bones are often so marked as to make identification 
possible from the pelvis alone. 

Recent studies on bone density reveal marked differences be- 
tween American Negroes and American whites. The weight/ 
volume ratio is higher in Negro skeletons indicating a greater 
degree of mineralization. This difference is in the opposite direc- 
tion of what might be expected on a purely nutritional basis. 




Many racial differences in the dentition are known and docu- 
mented, due in part to the fact that the teeth are equally acces- 
sible to study in the living and in skeletal collections as well. 

Tooth size varies considerably from population to population, 
the largest teeth being reported for Australian aborigines, and the 
smallest for the European Lapps. In the former, the crown length 
of the first molar averages 13.0 mm. (in the male) as compared 
with slightly less than 11 mm. in the Lapps. Notably, there are 
marked differences in tooth size within each geographical race. 
For example, there are both small-toothed and large-toothed local 
races in Africa and the teeth of Greenland Eskimos are larger 
than those of the Aleut. 

The form of the teeth also differs from group to group ( Fig. 6 ) . 





100 -i 


"a5 ~ 


2 80- 





•t= 60- 





c/> — 














H— — 









O _ 







i 1 



1 1 













Fig. 6. Frequency of Carabelli's (accessory) molar cusp in different groups 

showing maximum frequencies in Europe. Where frequencies vary the 

ranges are indicated. (Data from Carbonel, V. M.: The tubercle of Cara- 

belli in the Kish dentition. /. Dent. Res., 39:124-128, 1960.) 


Among many Asiatic populations the third molar roots are fused, 
whereas this condition is less common in Europeans and least so 
in many Africans. The number of cusps on the posterior teeth is 
similarly variable, with cusp reduction evident in both European 
and Asiatic peoples. The most impressive racial variation in the 
form of the teeth, the "shovel-shaped" trait, is especially well 
developed in some American Indian tribes. Whereas the back 
surface of the incisors is slightly "shovel-shaped" in about 15% 
of Finns, the incisors may be rolled in almost tubular shape in 
Pima, Navaho, Hopi and other Indians, in whom the shovel shape 
may extend to the canines as well ( cf . Fig. 7 ) . 

Finally, agenesis ( congenital absence ) of the third molars, rare 
in most Africans and verified in perhaps 6% of Europeans, may 
reach 30% in some Amerindian, Eskimo and Asiatic groups. 


Since the rate of growth and the timing of maturation are both 
greatly affected by the caloric reserve, few definite statements 
about racial differences in growth progress can now be made. 
Popular notions often turn out to be wrong, as for example, the 
supposition that children from southern or darker-skinned races 
mature earlier. 

Nevertheless, some racial differences in growth are clear-cut, 
among them differences in the absolute and relative growth rates 
of the limb bones in American Negroes, Japanese, and American 
"whites." Here, the characteristic differences in leg proportions 
are well established during fetal life, as Adolph Schultz (1926) 
has shown. 

Another racial growth difference involves the age at calcifica- 
tion of some of the bony nuclei of the wrist, foot and leg which ap- 
pears to be earlier in American Negroes and in selected African 
populations. This may represent a simple difference of patterning, 
or it may be associated with osseous and neuro-muscular advance- 
ment during infancy as Dean has demonstrated. 

For the teeth, however, the data appear to be incontestable. 
Almost every group averages earlier in the eruption of the per- 
manent dentition than do Europeans or Americans of European 
origin. Even with the disadvantages of malnutrition, disease or 







•>x-v»: ASIATICS 





■ 18% 

wi5% \Hypothenar 

:•>:•:•: 15% 

Fig. 7. Three areas of racial differentiation. A, the shovel-shaped trait on 
the posterior aspect of the incisors, especially common in the American 
Southwest. B, the inner or "Mongoloid" eyefold as contrasted with a typical 
European eye. C, geographical differences in specific palm-crease pattern 
frequencies documenting Amerindian-Asiatic differences in the thenar-first 
interdigital and the third interdigital areas. (From Rife, D.: Dermato- 
glyphics as ethnic criteria, Am. J. Hum. Genet., 6:319-327, 1954.) 


capture and life in concentration camps, Navaho, Maya Indian, 
Formosan, American Negro and Aleut children are ahead of norms 
for British and American "whites." Third molar eruption, rare 
before age 17-20 in most Americans, has been reported as early 
as the twelfth year in Kenya natives. These differences in tooth 
eruption are probably paralleled by racial differences in tooth 
calcification, a suggestion supported by a preliminary analysis of 
one American Indian group. 


Investigations in human physiology and biochemistry have re- 
vealed a great many apparent racial differences, but in most cases 
the purely genetic nature of these differences have as yet not been 
confirmed. Thus, most "natives" have lower blood pressures and 
lower serum cholesterol levels than is true for individuals of 
western European origin, but both blood pressures and cholesterol 
levels rise with urbanization and westernization. "Racial" differ- 
ences in the basal metabolic rate after correction for temperature 
appear to be largely functions of body build. Much more con- 
vincing are differences in drug-sensitivity (chapter VII), ab- 
normal haemoglobins and blood groups discussed in later chap- 
ters, and the haptoglobins to be mentioned shortly. 

Differences in the amount of excreted 17-ketosteroids do seem 
to be well-documented with considerably lower excretion levels in 
South African natives, as measured both in Africa and England, 
citing the work of Barnicot and Wolffson (1952). This does not 
necessarily mean that Africans are lower in the production of 
androgenic steroid hormones. It may mean that, through differ- 
ences in either adrenal-cortical activity or liver metabolism they 
excrete less of the C-17 steroids and more of the C-20 steroids not 
measured by the reagent used. Male pattern balding may also be 
considered along with 17-ketosteroid excretion levels, but as an 
example of hormone target-organ relationships. Since American 
Indians do not develop male pattern balding following testos- 
terone administration, their resistance to balding may be attrib- 
uted to target-organ insensitivity rather than to hormonal in- 
sufficiency ( cf . Hamilton, 1951 ) . 

Best documented to date, of the apparent racial differences in 



urinary excretion patterns, is the excretion of beta-amino isobu- 
tyric acid (BAIB) in Japanese, Chinese and Amerindians. Rarely 
excreted in adult Europeans, this is a common excretion product 
in Orientals tested both in Asia and in the United States. BAIB 
excretion is also found in Chinese and Japanese in America sub- 
sisting on purely occidental foods ( see Fig. 8 ) . So far BAIB holds 
the same position among the biochemical traits that the Diego 
factor holds among the blood groups, straight, coarse hair among 
the hirsute traits, and the suture or Wormian bones hold among 
the osseous differences. All are suggestive of "Asiatic" origins. 


In addition to marked racial differences in the frequencies of 
the various blood group alleles (Chapter IV), and in the fre- 
quency of the "abnormal" haemoglobins (Chapter VI), there is 
increasing evidence as to the differential distribution of the gene- 
determined haptoglobin "types." 

The haptoglobins, haemoglobin-binding proteins, are stable 
proteins, well differentiated by electrophoretic analysis of blood 
sera. There are at least three haptoglobin types Hp 1-1, Hp 2-1 
and Hp 2-2, ( Fig. 9 ) apparently controlled by the genes Hp 1 and 








Fig. 8. Diagram of a urinary chromatogram showing the area occupied by 

amino acids including BAIB (^-amino isobutyricacid). This amino acid 

is rarely excreted by Europeans and Africans, but frequently by Asiatics and 

Micronesians. Compare with frontispiece ana see text. 



Hp 2 . A person heterozygous for Hp 1 and Hp 2 therefore exhibits 
the haptoglobin type Hp 2-1. 

The Hp 1 gene, the less common allele in Europe, has a fre- 
quency of about 0.36 in Finns, Norwegians and Basques. It is 
more common in American Negroes (gene frequency approxi- 
mately 0.41 ) . As might be expected, therefore, the Hp 1 gene is the 
predominant haptoglobin allele in some parts of Africa with a 
gene frequency of 0.59 in Nigeria and in excess of 0.73 in Liberia. 
The Hp 1 gene is far lower in the Bushmen, according to a recent 
report. * 


Racial differences in dietary practices have long been viewed 
with interest and the suggestion has been made that people who 
prefer hot, spicy foods differ in taste-acuity from those who 


Starch gel 

Starch block 












6 8 10 
5 rate 

Fig. 9. The three haptoglobin phenotypes Hp 1-1, Hp 2-1 and Hp 2-2 as 
shown by starch-gel electrophoresis, starch block electrophoresis and by the 
centrifugation of blood serum. (From Beam, A. G., and Franklin, E. C: 
Some genetical implications of physical studies of human haptoglobins, 
Science, 110:596-597, 1958.) 

* Barnicot, et al: Nature, 184: 2042, 1959. Both in the Americas and the New 
World the frequency of the Hp-1 phenotype increases toward the Equator. 


espouse the cult of culinary blandness. So far, however, conclu- 
sive evidence as to racial differences in the classic taste parameters 
( sweet, sour, salty and bitter ) has not been achieved though there 
are gene-determined differences in sensitivity. 

Nevertheless, there are individual differences in the capacity to 
taste phenylthiocarbamide ( PTC ) and a large number of related 
substances, all of which have some antithyroid activity. These 
differences are apparently controlled by a pair of allelic genes T 
( for tasting) and t (non-tasting): the genotypes TT and Tt are 
tasters, and the tt genotype corresponds to the non-taster pheno- 
type. However, there is an hormonal effect as well, as evidenced 
by the greater proportion of taste-sensitive women. From the 
work of Fischer and Griffin (1959) taste-sensitivity may be ex- 
plained by the genetically-determined level of di-iodo tyrosine in 
the saliva. 

Taste sensitivity varies from population to population. Among 
American Indians the majority are tasters. In fact some Amer- 
indian populations must approach 1.00 for the taster (T) gene. 
In Africa and the Middle East the tasters are in the majority. 
However, in numerous parts of Europe the gene frequency for T 
is considerably lower and there are up to 43% of non-tasters in 
India (Fig. 10). 

Boyd has suggested that tasting is adaptive in some areas of the 
world, and inadaptive in others. At the present time the world 
distribution of PTC-propylthiouracil tasting not only evidences 
marked race differences but suggests many experiments bearing 
on PTC-taste sensitivity. "Tasters" for example, probably have 
more food aversions than non-tasters, a quality of limited value 
in times of scarcity. 


Race differences exist throughout the body and into the area of 
metabolic activity and biochemical functioning. Not only are 
there race differences in the pigmentation of the skin, eyes and 
hair, and in the morphology of the lips, nose, eyelids and mouth, 
but there are also differences in the inner organs, in the muscles 
and in the patterning of subcutaneous fat. The teeth, hemoglobins 
and hemoglobin-fixing proteins, taste acuity, drug sensitivity, 
urinary excretion patterns and probably sex-hormone activity 




6 52 

5-i i^ - 


£ o 

S o 

a. | 

.2 BQ 


^ 2 




£3 S 




■*-' T3 



3 < 

■K W 





* • 

cd bJO 

a § 

lity to ta: 











3 "° 

2 o> 



So 2 


4-> P CD • 



t§ * 



s ^ 

53 "£ 5> 


"° HH 

T3 W 




S .h 





3 s 



CD oo 





lison, A. C 
mong Alas 

■5 O 

3 * 



Eh O 



cd «y 

■■5 >-i 

Ph o 

o *a 

£ 1 


— &l 



exhibit measurable differences in averages or proportions from 
race-population to race-population. To say that "race is only skin 
deep," a statement notably but inaccurately repeated in many 
textbooks, is patently naive in the extreme. 

Some of the race differences are so marked as to allow little 
overlapping between geographical races. By way of example, 
the extreme spiral-tuft form of the head hair is virtually unknown 
in Europeans. The Rh negative gene, on the other hand, is rare 
outside of Europe. For many other traits, it is the frequencies or 
proportions that are distinctive, as with the haptoglobin types in 
Europe and Africa. Often, differences are of a small order of 
magnitude, large enough to attain statistical significance but not 
impressive enough to use in a taxonomy. 

Not infrequently, particular differences within geographical 
races are more marked than those between geographical races. For 
example, the phenomenon of steatopygia ( fat rears ) virtually sets 
off the Bushmen-Hottentot from the rest of Africa, and in body 
build the Papago are distinct from other Amerindian groups. But 
a single such difference probably reflecting local selective forces 
does not alarm the taxonomist, no more than (and for the same 
reasons) that a run on the bank does not affect the long-term 
position of a gambling casino. 

It is important, however, to emphasize the independence of 
genes. The fact that genes P and Q are both common in a given 
group does not mean that they are inherited together, that they 
came from the same source. Whereas the PQ genotpye may be 
characteristic of a given population, individuals who are pp or 
qq are not less "pure" on that account. And since the genes are 
independent, the existence of the gene P in a given population 
need not be evidence that it came from a PQ source. In other 
words, dark skins are not necessarily of "Negro" origin, nor in- 
ternal eye-folds inevitably due to "Mongoloid" ancestry. Failure 
to recognize the independence of genes has led to some rather 
implausible historical reconstructions in the past. 

Races differ in a great many gene-determined respects and a 
marked difference in the proportion of one set of alleles predicts 
nothing in respect to the proportions of another set. Moreover, 
it is not the mere fact of difference that intrigues us today, but 


rather the source of the difference. The fact that one race has a 
high frequency of some trait, and there is a low frequency in 
another race, is mildly interesting as a descriptive fact but no 
more until we turn our attention to the reasons why. 


Racial differences are known to exist in almost every area of 
anatomy where comparative data have been accumulated, and 
there is growing evidence for race differences in biochemical 
functioning and in the constituents of cells and tissues. 

In a minority of examples, there may be little overlapping 
between geographical races. Much more frequently, differences 
are merely matters of proportion, the incidence of different traits, 
or the frequencies of the allelic genes that determine them. Not 
infrequently a given gene-determined trait varies more extensively 
among local races in the same geographical race than between 
geographical races. 

The independence of the genes that make for similarities in 
differences must be clearly understood. Thus two races may be 
markedly different with respect to one allelic pair of genes and 
not at all different with respect to another. Taken alone, there- 
fore, a particular set of differences ( or similarities ) says nothing as 
to communality or divergence of origin or descent. 


Barnicot, N. A. and Wolffson, D.: Daily urinary 17-ketosteroid output 

of African Negroes, The Lancet, 262:893-898, 1952. 
Fischer, R. and Griffin, F. On factors involved in the mechanism of 

"taste-blindness," Experientia, 15:447-451, 1959. 
Garn, S. M.: Types and distribution of the hair in man, in Hamilton, 

J. B. (ed.) The Growth, Replacement and Types of Hair, Annals 

of the N. Y. Academy of Sciences, 53:498-507, 1951. 
Garn, S. M. and Moorrees, C. F. A.: Body -build and tooth emergence 

in Aleutian Aleut children, Child Development, 22:262-270, 1951. 
Gartler, S. M., Firschein, I. L. and Kraus, B. S.: An investigation into 

the genetics and racial variation of BAIB excretion, Am. J. Human 

Genet., 9:200-207, 1957. 


Gartler, S. M., Firschein, I. L. and Gidaspow, T.: Some genetical and 

anthropological considerations of urinary /?-aminoisobutyric acid 

excretion, Acta Genetica, 6:435-466, 1956/7. 
Hamilton, J. B.: Patterned loss of hair in man, types and incidence, 

in Hamilton, J. B. (ed.) The Growth, Replacement and Types of 

Hair, Annals of the N. Y. Academy of Sciences, 53:708-728, 1951. 
Hoyme, L. E.: Genetics, physiology and phenylthiocarbamide, J. 

Hered., 46:167-175, 1955. 
Lasker, G. W.: Genetic analysis of racial traits of the teeth, Cold 

Spring Harbor Symposia on Quantitative Biology, 15:191-203, Cold 

Spring Harbor, The Biological Laboratory, 1950. 
Lee, M. M. C. and Lasker, G. W.: The sun-tanning potential of human 

skin, Human Biol, 31:252-260, 1959. 
Makela, O., Erikson, A. W. and Lehtovaara, B.: On the Inheritance 

of the haptoglobin serum groups, Acta Genetica et Statistica Medica, 

9:149-166, 1959. 
Martin, R. and Sailer, K.: Lehrbuch der Anthropologic, Stuttgart, 

Gustav Fischer, 1959. 
Moorrees, C. F. A.: The Aleut Dentition, Cambridge, Harvard Uni- 
versity Press, 1957. 
Moorrees, C. F. A.: Torus mandibularis : its occurrence in Aleut chil- 
dren and its genetic determinants, Am. J. Phys. Anthropol, N.SJO: 

319-330, 1952. 
Schultz, A. H.: Fetal growth of man and other primates, Quart. Rev. 

Biol, 1:465-521, 1926. 
Steggerda, M. and Hill, T. C: Eruption time of teeth among Whites, 

Negroes and Indians, Am. J. Orthodont. and Oral Surg., 28:361-370, 

Sutton, H. E. and Clark, P. G.: A biochemical study of Chinese and 

Caucasoids, Am. J. Phys. Anthropol, N.S. 13:53-66, 1955. 
Trotter, M., Broman, G. E. and Peterson, R. R.: Density of cervical 

vertebrae and comparison with densities of other bones, Am. J. Phys. 

Anthropol, N.S. 17:19-25, 1959. 



Jlor many centuries the possibility of transfusing 
blood has appealed to surgeons. Great numbers of men died from 
loss of blood following accidents or battle. Patients expired during 
and following operations, patients who could have been saved by 
blood transfusion. But until 1900 transfusions were impractical: 
too often they ended in shock and death. Somehow, blood just 
didn't mix. 

Then, about fifty years ago, Moss and Jansky discovered the 
existence of four different blood types which they named types 
1,11, III and IV. By determining the blood "type" of the patient, 
and transfusing compatible blood, blood transfusions became a 
fortunate practicality. And, as a result of the extensive work on 
blood-typing, during and since World War I, a tremendous 
amount has been learned about blood types and their relation 
to local and geographical races. 


The original four types of Moss and Jansky have since been 
renamed A, B, AB and O. These blood types are simply inherited 
with both A and B dominant over O. Genetically, type O blood 
is OO, while the phenotypes A and B may be either AA or OA, 
and BB or OB respectively. 

Blood Type 

Possible Genotypes 









Of the three blood types (or factors) in the A-B-O system, O 
is the most common the world over. Among some American 



Indian groups, over 90% of the people are of blood group O, with 
gene frequencies for O as high as 0.99. In much of Europe, the 
incidence of O is from 35% to 40%, and the frequency of the gene r 
is between 0.6 and 0.7. Among Chinese, Japanese and Asiatic 
Indians and in many African populations, O may be as low as 30% 
with a gene frequency of as little as 0.5. However, even within a 
particular geographical race there are marked differences in the 
incidence of O. By way of example, O has been reported in 97% 
of Utes, but in only 23% of Blackfeet Indians. In Europe there is 
an increasing incidence of blood type O from south to north, a 
trend that is repeated even within the British Isles. 

Blood group A, the next most frequent blood type the world 
around, similarly evidences its own pattern of distribution. It is 
rare (under 5%) in some Amerindian populations, yet extremely 
common (over 75%) in others; both are world extremes, and point 
to the wide divergences possible between local races. By way of 
comparison, the phenotype frequency of A is about 45% in Eng- 
land (with a gene frequency of 0.25), and the same among 
Americans of Northwest European descent. 

Blood group B, the least common of the three types in the 
A-B-O system, has the most interesting world distribution. It is 
completely absent in most North and South American Indians, 
rarely over 2% in others, and this may be due to admixture. Blood 
group B is less common than A in Europe, with a phenotype fre- 
quency of 9-25%, averaging about 15%. However, B which in- 
creases in frequency to 22% in Ukrainians and 25% in Egyptians, 
reaches maxima of 35-37% in China, Java, Bengal and the Siberian 
north. B is an Asiatic and African blood type much more than it 
is European and it is not at all Amerindian. 

Comparing Sumatra, Java and the Philippines with Polynesians, 
there is a great divergence in A-O-B blood groups. The Asiatic- 
Malaysian areas are high in B, whereas Polynesia is low in B 
(under 3%). Clearly Polynesians cannot be derived from any 
recent mixture with Malaysian people. In similar fashion, the 
low to zero incidence of blood group B in aboriginal America 
precludes major recent contact with Asiatic Mongoloids whose 
frequency of B ranges from 20 to 40% in different populations. 

However, blood type A is divisible into two subtypes, Ai and 


A 2 ; of the two Ai is far more common than A 2 . The incidence of 
A 2 varies from zero (in Amerindians and Australians) to 10-15% 
in much of Europe. American Indians agree with Asiatic Mongo- 
loids in the virtual absence of A 2 , as is true of Polynesia and 
Australia as well. In fact, A 2 is practically limited to Europe and 

Broadly, the A-B-O blood system can be summarized by noting 

(1) the near absence of B and absence of A 2 in the Americas, 

(2) the low to moderate frequencies of B and increasing fre- 
quency of A 2 in Europe, and (3) the high incidence of B and 
relative rarity of A 2 in Asia. But the Australian aborigines seem 
practically Amerindian, as do the Polynesians in their low inci- 
dence of B and absence of A 2 . The A-B-O system, taken alone 
would suggest a separate origin for them, whereas it groups Asia, 
India and Africa. 


Next in order of discovery to the A-B-O system, and therefore 
in the amount of information we now have, is the MNS-U system. 
For years it was the M-N system, then S was added, and more 
recently U ( sometimes written S u ) . 

As with A and B, M and N are inherited without dominance, 
A person may be M, N or MN, corresponding to the genotypes 
MM, NN and MN. For much of the world, the frequencies of M 
and N are about equal, and rarely is either M or N entirely 
missing. Thus, in England the gene frequencies for M and N are 
close to 0.53 and 0.47 and in Japan they approximate 0.56 and 0.44 

But M is peculiarly high among the American Indians, and in 
many Amerindian tribes there is little or no N. ( Low values for 
N have also been reported for the Near East ) . In contrast to the 
situation in America, Australia is the virtual homeland of N, and 
M — that is MM individuals are entirely absent in some aboriginal 
hordes. Throughout the Pacific, in Papua, Fiji and Hawaii blood 
type N is similarly predominant over M. Whereas these peoples 
of the Pacific are similar to Amerindians in the low incidence of 
B and of the subtype A 2 , the high N frequencies of Pacific people 
set them off completely and preclude recent major contact with 
America, as their absence of B does with Asia. 


Since the extremes of M and N are found in populations 
formerly explained on the basis of "admixture" it is notable that 
no combination of Asiatics could yield the low N values common 
in America, and no Caucasoid-Negroid combinations could yield 
the nearly M-free peoples of Australia. Either the "original" three 
races never existed, or subsequent evolution has so altered their 
genetic makeup in particular localities as to make proof of the 
three-race hypothesis quite impossible. 

Still, two additions to the M-N system must be made. One 
involves the rare gene S, discovered in 1947, which involves a 
mutation alternately from M or from N. A person may be MS, 
MNS or NS. The S gene, discovered in England, (where it is 
quite common) has since been shown to be absent in Australian 
aborigines, but present in natives of New Guinea. Tentatively, 
therefore, one may suggest either closer affinities between 
Papuans and Europeans, or an absence of ties between the Aus- 
tralian and European geographical race. Perhaps MS and NS 
might be called M 2 and N 2 , in which case we could say that both 
M 2 and N 2 are fairly common in Europe, but absent among the 
aboriginal Australians. 

The second addition to the MNS system involves the gene U, 
sometimes written as S u . Among Europeans everyone is U-posi- 
tive. In Negroes, or more specifically the American Colored, about 
1% are U-negative as shown below. This polymorphism leads to 
complications, both in transfusions and in childbirth. Transfusing 
U-positive blood into a U-negative Negro can be dangerous. 
While segregated blood stores would not be the answer (most 
Negroes are U-positive as are nearly all Europeans ) the necessity 
to consider race in planning transfusions becomes all the more 
apparent. Further divergences in Negro blood groups will be 
mentioned later in this chapter. 


Milwaukee "Caucasians" 
Milwaukee Negroes 
New York Negroes 

From Green wait, Sasaki, Sanger and Race (1956) and Sampson and Thomas 



















By now, most educated individuals have heard about the 
Rhesus factor and Rh incompatibility. They know that, especially 
in later pregnancies, a rhesus-positive fetus can be damaged by 
antibodies produced by the Rh-negative mother. 

Actually, there is a long series of rhesus genes, Ri, R 2 , Ro, r', r 
variously written as R', R", etc. Practically, Ri, R 2 , f and others 
are dominant over the rhesus negative gene, which will therefore 
be written as r;* the rhesus-negative individual is a homozygous 

Among Europeans in whom the rhesus-negative gene was first 
discovered, the proportion of Rh negative individuals ranges from 
about 12% through an average of approximately 15% in England 
and the United States, to nearly 30% in Basques. 

Elsewhere in the world, rhesus-negative blood is rare, un- 
common or even totally absent. A long series of American Indians 
and many series of Papuans or Australian aborigines can be 
compiled with few or no rhesus-negative individuals. Rhesus- 
negative blood is also absent in Polynesia, and uncommon in 
China, Japan and the Philippines, though the incidence begins to 
rise in India, Pakistan and Afghanistan. 

Among American Negroes the type rr is about half as common 
as in northwest Europeans, leading expectably to its relative 
rarity in the sections of Africa whence their African genes origi- 
nated. However, the type Ro, which rarely exceeds 2% in Europe, 
achieves maxima of more than 40% in the American Colored, and 
over 70% in Africa. It is for this reason that the Ro gene in 
Melanesia is of interest. 

The present distribution of the "rhesus" blood types, the high 
frequency of the gene r in Europe and especially among the 
Basques, the rarity of r in Asia, Australia, Polynesia and aboriginal 
America, and the concentration of Ro in Africa, cannot be ex- 
plained in any simple way. One is loath to accept Boyd's "Early 
Europeans," who presumably contributed vast numbers of the 

* The notation for the Rh alleles used here is primarily that of Alexander Weiner 
though it does not follow his more recent revisions. Abroad there is a totally different 
notation developed by R. A. Fisher, which involves different assumptions about the 
mode of inheritance. 


rhesus-negative genes in Europe and were then absorbed by late- 
comers. Even if there had been such a population, we are still 
faced with the continued survival of this disadvantageous gene, 
the more so under primitive conditions. And we wonder what 
value Ro has in Africa. 


In 1950, a new blood-group factor, unrelated to the A-B-O, MN 
and Rh systems, was found in the blood of a Mr. Duffy, and the 
new system that eventuated from this chance event was named 
after him. The Duffy factor soon proved important in trans- 
fusions, as a cause of transfusion reactions. 

Briefly, there is the Duffy-positive gene (Fy a ) and the Duffy 
negative allelomorph ( Fy b ) . A person may be Fy a Fy a , Fy b Fy b or 
Fy a Fy b . However, since Fy a is dominant over Fy b , only two 
phenotypes are identifiable. Therefore, phenotype frequencies of 
Fy b are used in computing the gene frequencies for Fy b and Fy a 

In England 65% of subjects proved to be Duffy-positive, cor- 
responding to a gene frequency of 0.40. Far higher gene fre- 
quencies for Fy a have appeared in Pakistan, India and among 
New York Chinese, and Australians. Far lower gene frequencies 
were observed in American Colored individuals ( see below ) . 

(Fy a )GENE 

Group Studied 

Gene Frequency Fy a 

Cape York Australians 








East Indians 


American Indians 


English donors 


Minnesota Whites 


American Colored 


From Matson and Swanson (1959), Simons et al. (1958) and Race and Sanger 


Obviously, the home of the Duffy-positive gene is in the Pacific 
and Eastern Asia with decreasing frequencies both southward into 
the Americas and westward into Europe and then Africa where 
the Fy a gene is virtually absent. The distribution of Fy a , if taken 
seriously by proponents of a simple three-race theory, would 
provide better evidence that the Europeans are of mixed Poly- 
nesian-Negro origin (since they are intermediate in Fy a fre- 
quencies) than for any tri-hybrid origin for the Polynesian 


The Diego blood group system, one of the most recently dis- 
covered, involves a pair of genes Di a and Di b , and the two 
pheno types, Diego "positive" and Diego negative. Though Diego- 
positive individuals are nowhere in the majority, the Di a gene 
clearly separates Australia and the Pacific from Asia and the 
Americas (Fig. 11). 

In two series of Australians, one in the north nearest New 
Guinea, and one in Central Australia, no Diego-positive individ- 
uals were found. Similarly the Diego antigen appears to be absent 
in the eastern Polynesians. On the other hand, Diego-positive 
individuals have been reported in 25% of Peruvian Indians and 10% 
of Penobscot Indians, but rarely among Alaskan ( Tlingit ) Indians 
and Alaskan Eskimos. Several studies confirm the virtual absence 
of Di a in Alaska, and the relatively high frequencies in Central 

Despite the absence of the dominant Di a gene in Alaska, Diego- 
positive individuals are frequently found among Chinese, Japa- 
nese and Koreans. This suggests, quite reasonably, an Asiatic 
source of Diego in the Americas. However the extremely variable 
phenotypic frequencies in Central and North America, ranging 
from 0% to over 20% and the absence of Diego nearest the Asiatic 
mainland can be explained only in terms of local selection. 




20 40 60 

» ' I I L_J 

South American Indians 



North American Indians 

Southeast Asiatics 


African and American Colored 

Australian and Oceanic 





Fig. 11. Percentage of Diego-"positive" individuals in different populations. 
The genotype Di a or Di a+ is largely restricted to Amerindians and Asiatics. 
(Adapted from Layrisse, M., Wilbert, J., and Arends, T.: Frequency of 
blood group antigens in the descendants of Guayqueri Indians, Am. J. Phys. 
Anthropoid N.S. 16:307-318, 1958.) 




Blood Group System Description of Phenotype Frequencies 


(including Ai & A 2 

(or S u ) 


(Ra R 2 R r' r etc.) 


(Fy a Fy b Fy) 

(Di a Di b 


(Jk a Jk b ) 

O most common group, with over 50% of individuals in 
most populations of this type. B nearly absent in Aboriginal 
America and Australia, progressively more common in 
Europe (15%), Africa, India and Asia (up to 40%). A-, 
practically limited to Europe. 

American Indians almost exclusively M, N most common 
in Australia and the Pacific. MS and NS absent in Aus- 
tralia. U-negative rare, apparently limited to Africa.* 

Rh negative individuals (rh) rare or absent in most of 
the world, but approximate 15% in Europeans. Of the 
Rh positive alleles (Ri, R a etc.) the R form is primarily 
found in Africa (up to 70%). 

Most Australians and Polynesians§ and 90-99% of Asiatics 
Duffy-positive (Fy a ), 90% in India, 85-90% of most Ameri- 
can Indian,** 65% in England and America, 27% in 
American Negroes. Fy a very low in Africa but the gene Fy 
extremely common (>80%).f 

Diego-positive (Di a ) individuals limited to Amerindians 
(2-20%), and Asiatics.* Di a absent in Europe and Africa, 
Australia, Micronesia and Polynesia,§ and in Eskimos.* 

Kidd positive (Jk a ) most common in West Africa and 
American Colored (>90%), North American Indians 
(70%-90%), Europeans (approximately 70%), and least 
common in Chinese (50%-55%).j ** 

Special References: * Greenwalt et al. (1958), f Race and Sanger (1954). 
mons, Graydon and Gajdusek (1958), ** Matson and Swanson (1959).* 



As a result of numerous investigations involving millions of 
people, it is extremely clear that the blood groups are subject to 
natural selection. Population differences in blood group frequen- 
cies may therefore be viewed as the product of competing lines 
of selection, the balancing of selective advantages and disad- 
vantages associated with each serological type. 

Classic Rh incompatibility, the loss of a Rh heterozygous infant 

* Levine has shown a protective effect of certain ABO genetypes against Rh 
incompatibility. See Human Biol, 30:14-28, 1958. 


due to antibody formation in a Rh-negative mother, is the best 
known example of selection against a blood type. The long-term 
effect of such incompatibility would be to eliminate the Rh" 
gene (r) from the population, and this may explain the rarity or 
absence of the Rh negative gene in much of the world. At the 
same time we need some explanation for the continuance of this 
gene in Europe, especially among the Basques. 

Less well known, but equally important, is maternal-fetal in- 
compatibility in the A-O-B system. Since the net effect would be 
to eliminate types A and B differentially in comparison to O, this 
may provide an explanation for the numerical predominance of O 
the world around. Moreover, now that transfusion reactions have 
been discovered for a variety of other blood group factors, includ- 
ing the rare U in the MNS-U system it is obvious that the increas- 
ing use of blood transfusions is serving to eliminate some propor- 
tion of the less common genes in each population. 

Apparently, serological incompatibilities are one cause of in- 
fertility. The A-B-O phenotype frequencies proved markedly ab- 
normal in a series of infertile couples studied in Michigan. Since 
such incompatibilities are not likely to occur in O couples, there 
is obviously a slight adaptive advantage to the OO genotype. 
Reproductively, at least, the type O male is a "universal donor." 

Recently, associations between A-B-O types and chronic ill- 
nesses have been demonstrated, among them an association be- 
tween type O and ulcers and type A and gastric cancer. Although 
criticizied on statistical grounds and challenged outright by 
Weiner, similar results have been obtained in Europe, America 
and Japan. However, since the diseases involved occur late, 
commonly after the reproductive period, the importance of this 
kind of selection remains open for study. 

In view of the fact that the blood groups are subject to selection 
the use of serological data in historical reconstructions is obviously 
limited. It would be unwise to guess as to the frequency of B in 
Asia ten thousand years ago, or even the frequency of the Rh 
negative gene among "early" Europeans. However, since all lines 
of selection now known would tend toward population homozy- 
gosity, special attention should be given to situations where, as 
with the Duffy factor in Europe, maximum number of heterozy- 


gotes exist. There may well be a real adaptive advantage to 
being MN, Fy a Fy b , and possibly AB. In fact, there are more MN 
children from MN x MN marriages than one might expect even 
if technical errors are to some extent involved. 


The blood groups are simply inherited and therefore qualify as 
ideal traits for use in racial comparisons. Blood group determina- 
tions can be made with relatively high reliability except for the 
less common antigens ( Osborne, 1958 ) , and this commends them 
for general use. Besides, blood group determinations are made 
by the millions every year in the course of blood donating and 
transfusions, thus providing a wealth of free information on blood 
groups in different races. 

All of these advantages were clearly recognized over thirty 
years ago, and the idea of a serological taxonomy was advanced. 
After all, blood groups are genetical, accurate, stable and sus- 
ceptible to statistical analyses. A serological taxonomy made at 
least as much sense, and in fact more, than a classification based 
on hair form or skin color. 

But the first serological taxonomies were hardly reasonable. 
Using the A-B-O system, Asia, India and Africa ended up in one 
pile ( based on the frequency of B ) , and aboriginal America and 
Australia in a second sorting ( based on an absence of B ) . Totally 
unrelated populations were characterized by similarities in A-B-O 
frequencies and contiguous groups compared serologically were 
practically in separate planets. 

The situation improved somewhat with the discovery of the 
subtypes of A, which belatedly confirmed Amerindian-Asiatic 
similarities and no longer made Europe an apparent Asiatic- 
Australoid mixture. M-N data further appeared reasonable, geo- 
graphically as well as genetically. The most unusual distributions 
of M and N were geographically most isolated from Europe, Asia 
and Africa. 

With the discovery of the Rhesus system of alleles, serological 
and geographical taxonomies became more nearly reconciled. 
Africa appeared clearly separable from Asia (despite high fre- 
quencies of B) and Europe from both. Even so, considerable 


juggling was necessary to make serological taxonomies coincide 
with natural populations. What had happened to the serologists 
was exactly what had happened to morphologists. Using the 
conventional systems of differences they had confused their 
criteria ( here the blood groups ) with the races they were trying 
to describe. By near-sightedly working with the gene frequencies 
and ratios, they had come up with an artificial classification, a 
system of blood groups rather than a classification based on 
natural populations. To the extent that serological classifications 
worked, they were attempts to describe natural populations (as 
Boyd has done in his Boston University Lecture).* When the 
classifications were based on serological criteria, rather than the 
populations, the results were understandably bizarre. 

Actually, the major contribution of the blood groups is not the 
establishment of a taxonomy, but to the more adequate compari- 
son of related race-populations, their similarities and differences. 
By way of example, the virtual absence of B in Australia (except 
in the Cape York area) and the absence of S (that is MS and NS ) 
confirms the separation of Australian and Papuan peoples. At the 
same time, the absence of Diego and the high frequencies of N, 
align the Australians with other peoples of the Pacific. Similarly, 
the similarities between Amerindians and Asiatics in the Duffy 
factor and in Diego, are balanced by the differences (in blood 
group B and possibly in the Kidd blood group ) . 

Particularly for populations formed by recent admixture, the 
blood groups afford precise quantification. Using the Rh-negative 
gene (r), the R gene, the incidence of the U-negative trait, Duffy 
(that is Fy a ) and the Kidd-positive trait, very exact estimates of 
the degree of intermixture can be obtained for the American 
Colored and Cape Colored populations, and for various tri-racial 
hybrids. Similarly, extensive incorporation of Amerindian genes 
into the American Colored population can be ruled out by sero- 
logical data. With such information on hand, deviations from 
expected proportions of various gene-determined traits, such as 

* Both in his earlier book, and in his Boston University Lecture, Boyd's system 
of "races" properly involved serological descriptions of natural populations rather 
than serological races based on blood-group frequencies alone. 


the abnormal haemoglobins, may then be attributed to natural 

But the blood groups themselves are subject to selection. Un- 
questionably, there has been selection against the Rhesus-nega- 
tive gene, and against both A and B in favor of O, at least in Euro- 
peans. For these reasons, present gene frequencies do not provide 
a perfect indication of what they were in the past, and guessing 
as to proportions of groups entering into ancient admixture be- 
comes a most hazardous activity. Valuable as they are in the 
study of contemporary populations, the blood groups are of lim- 
ited value in solving ancient ties between races, but so are the 
conventional morphological traits. 


The serological factors (blood groups) discovered since 1900, 
have added tremendously to the scientific study of race. Simply 
inherited and susceptible to analyses as gene frequencies, the 
ABO, MNS, Rh, Kidd, Duffy, Diego, Kell and Lutheran systems 
of alleomorphs make possible detailed analyses and comparisons 
of contiguous local races, and help in reconstructing origins of 
recently-mixed populations. 

As with classifications based on morphological traits rather than 
on the populations themselves, artificial "serological races" add 
nothing to human taxonomy. The major use of blood groups in 
classification is in the comparison and analysis of natural popu- 
lations, and in the study of natural selection in contemporary 


* Allison, A. C: Aspects of polymorphism in man, Cold Spring Harbor 

Symposia on Quantitative Biology, 20:239-255, Cold Spring Harbor, 

The Biological Laboratory, 1955. 
Boyd, W. C: Genetics and the Races of Man, Boston, Little, Brown 

and Company, 1950. 
*Boyd, W. C: Genetics and the Races of Man, Boston University 

Lecture, Boston University Press, 1$58. 
*Garn, S. M.: Race and evolution, Am. Anthropol, 5.9:218-224, 1957. 

Readings on Race, pp. 187-189, 1960. 


Greenwalt, T. J., Sasaki, T., Sanger, R. and Race, R. R.: S u an allele 
of S and s, Bibliotheca Haemotologica, 7:104-106, 1958. 

Matson, G. A. and Swanson, J.: Distribution of hereditary blood 
antigens among the Maya and non-Maya Indians in Mexico and 
Guatemala, Am. J. Phys. Anthropol, N.S. i7:49-74, 1958. 

Mourant, A. E.: The Distribution of Blood Groups in Animals and 
Humans, Springfield, Thomas, 1956. 

Osborne, R. H.: Serology in physical anthropology. Technical prob- 
lems as revealed by repeated blood determinations in twins, Am. J. 
Phys. Anthropol, N.S. 16:187-195, 1958. 

Race, R. R. and Sanger, R.: Blood Groups in Man, Springfield, Thomas, 

Sampson, C. C., Thomas, C. and Griffin, S.: Isosensitization to the U 
factor, /. Am. Med. Assn., 1 71 .1203-1204, 1959. 

Simmons, R. T.: The Diego (Di a ) blood group: tests in some Pacific 
peoples, Nature, 179:970-971, 1957. 

Simmons, R. T., Graydon, J. J. and Gajdusek, D. C.: A blood group 
genetical survey in Australian aboriginal children of the Cape York 
Peninsula, Am. J. Phys. Anthropol, N.S. 16:59-77, 1958. 

* Reprinted in Readings on Race. 



An the Origin of Species, an epochal book published 
one hundred years ago, Charles Darwin described the principal 
mechanism of evolutionary change. Pointing out the inherent 
variability of all living species, and their tendency to depart infi- 
nitely from a central type, Darwin saw in natural selection the 
directive force behind each species, and smaller taxonomic units 
as well. 

Natural selection provides a mechanism for change within 
species and especially within races. With natural selection in 
operation each race undergoes continual change. Ultimately, the 
descendants scarcely resemble their ancestors, phenotypically and 
genotypically as well. Though the many races of mankind derive 
from a common source, they have come to differ widely, both 
among themselves, and from their first sapiens progenitor. 

Natural selection involves no mystery, no mirrors, no Lamarck- 
ian "will to change" and no Bergsonian elan vital. All of the 
climatic forces we see about us are potential selective agents 
bringing about, through genetic adaptation, differentiation be- 
tween races. Food is a selective agent by its very abundance 
favoring sheer fertility, or by scarcity, favoring smaller size and 
slower growth. So is disease a powerful selective agent, favoring 
in each generation those with superior immunity. The natural 
world is full of forces that shape races and make one race different 
from another. 

Thus, when we view a race, long-established in a particular 
climatic zone, habituated to specific foods and levels of caloric 
intake, and subjected to local diseases, our first inclination is to 
explore the immediate environment for selective factors respon- 
sible for local differentiation. Or, having caught sight of some 



unusual difference, we may choose to investigate it alone and to 
determine what adaptive advantage its presence or absence may 

However, since local environments are multitudinous ( and for 
the most part ill-explored), it becomes quite a task to assess the 
race-making potentiality of a bog here and a swamp there. In- 
dividual human differences, numbering in the thousands, present 
a comparably formidable task to analyze. How do we measure the 
evolutionary significance of the inner or "Mongolian' eyefold, 
the reasons for a given gene frequency among the Tlingit, or the 
reasons for peculiar Duffy and Diego frequencies among the 

Here the broad-gauge view comes into its own. Instead of 
trying to match each racial difference to its corresponding selec- 
tive advantage, we often search for broad generalizations involv- 
ing variables of climate or major somatic differences. We look for 
characteristics particularly associated with extreme heat or cold. 
We look for climatic variables differentially associated with body 
build or skin pigmentation. We watch for diseases that may 
explain the distribution of the abnormal haemoglobins. In short, 
and detective-wise, we seek for clues that may explain particular 
directions of racial differentiation, as well as the process of race- 
making in general. 


One major environmental difference is temperature, and with 
it the intensity of solar radiation. World extremes run from 130 °F. 
in the shade, to minus 60 and below, a range of 190° F. Usually, 
the extremes are found in very different regions, but in the desert 
soil-temperatures may drop from 125 °F. at midday, to below 
freezing the very same night. The same man may come near to 
blistering his feet at noon, and frosting his toes in sleep. This is 
true in the Kalahari, in the deserts of Australia and in the deserts 
of the American southwest. 

Solar radiation, contributing to the radiant heat load varies 
by a factor of nearly 10,000. An exposure meter may go off-scale 
in the desert and barely yield a reading in the Arctic night-day. 
It is true that houses, umbrellas, central heating and air-condi- 


tioned rooms mitigate the extremes for many of us now, but this 
is not true for all, and such protection against nature is relatively 
recent in mans million-year existence. When we read of death by 
sunstroke or by cold, we may well realize how much more strin- 
gent climatic selection was a hundred and a thousand years ago, 
and what a large proportion of the population faced climatic ex- 
tremes with relatively little environmenal protection. 

We in America and Western Europe have unlimited access to 
food. With no additional cost we could become circus fat-men, 
ingesting a daily 6,000 calories instead of our national averages 
of 3,000 to 3,500 calories a day. Overnutrition is clearly a major 
cause of death with us : as a population we are adapting to a food 
surfeit. But in many parts of the world, 1,500 calories a day (a 
weight-reducing value for us ) , is a feast, and 800 calories per day 
is unfortunately "normal" for many. Our breakfast of bacon and 
eggs and buttered toast, and coffee with cream provides as much 
meat as many people get in a week, and is comparable in fat 
to several days' food for them. An American adult drinks more 
milk in a few days than most African and Asiatic toddlers see in 
a year. As with temperature, extremes in food intake are great. 
Death by starvation is in fact far more common than death by 
heat or cold. 

Besides temperature and food, there are numerous infectious 
diseases that slaughter people by the billions. In Lincoln's time, 
Americans died in great numbers of cholera, typhus, malaria, 
diphtheria, scarlet fever, pneumonia, tuberculosis and dysentery. 
Quite recently, Eskimos succumbed to measles and mumps and 
whooping cough. In other parts of the world malaria, elephanti- 
asis and yaws, amoebic dysentery and snail carried Bilharzia are 
common causes of death. Little wonder that in some primitive 
areas today scarcely one infant in three lives to adulthood. And in 
each part of the world, a unique combination of pandemic diseases 
contributes mightily to race differences. 


Variations in skin pigmentation in man have obvious adaptive 
value. As most of us know from personal experience, even a little 
melanin in the outer layers of the skin is protective. Early in the 


season, before we have begun to tan, even short exposures to 
bright sunlight may result in discomfort. Later, with a good tan, 
we can gambol for hours in the August sunshine. 

Darker skins represent natural protection against the most 
damaging portion of the solar spectrum, the ultra-violet and espe- 
cially about 2200-2800 Angstrom units. Not only is more melanin 
protective against immediate damage to the deeper layers, but it 
has a long-term value as well. Skin cancer is more common among 
"whites" in Texas, than in more northern parts of the country, and 
there are conspicuous Negro-white differences in the incidence of 
skin cancer, even with correction for occupation. 

Balanced against the advantages of pigmentation in brighter 
climates, are possible disadvantages in duller climes. Vitamin D, 
essential to bone growth, is produced by the irradiation of ergos- 
terol in the skin. Heavy skin pigmentation could be disadvan- 
tageous where the D vitamin intake is low leading to childhood 
rickets and birth complications. 

Protection, on the one hand, and interference with vitamin-D 
synthesis on the other may well explain the north-south gradient 
in skin pigmentation from Finland and Denmark to the Equator, 
and the smaller but equally apparent gradient in Asia. In addi- 
tion, one must note the contribution of skin pigmentation to the 
amount of radiant energy converted into heat in the most super- 
ficial layers of the skin. With darker skin reflecting far less of the 
visible and near-UV portion of the solar spectrum, American 
Negro skin temperatures are higher, under standard conditions 
of comparison and the inner (rectal) temperatures are distinctly 
higher as shown by Paul T. Baker ( 1958 ) . This would raise the 
heat load to be dissipated, but could be advantageous where 
water supplies were adequate. In fact, the relatively lighter skins 
of the Bushmen and Hottentot may be viewed as a compromise. 


Among the many differences between races, variations in body 
size are especially conspicuous. In some groups males average 
close to six feet, and in other populations average male stature is 
nearer to five feet. The fat-free body weight of American males 
approximates 135 pounds ( some go as high as 190 pounds ) , while 


the comparable fat-free weight in other groups may average as 
little as 105 pounds. 

Large body size can be advantageous. It commands respect, 
it is helpful in wrestling and hand-to-hand lighting, and it is a 
useful adjunct in hunting big game. The bigger man can cover 
more territory, he is speedier, he can tackle bigger game and bring 
back larger cuts of meat. Not too surprisingly, the noted hunting 
peoples of North America and Europe have been tall on a world 
scale. Given large animals to hunt, size is adaptive. 

But size and massivity have their disadvantages. Larger size 
requires more calories, merely to keep alive, as Americans, Dutch- 
men and Englishmen learned in Japanese concentration camps. 
Larger size requires more calories to grow on so that the geneti- 
cally-large child is at a particular disadvantage when food is 
scarce. And the large man, while more efficient at heat regula- 
tion in cold weather, is less efficient in hot weather. 

For small size, the advantages and disadvantages reverse. Size 
is of no advantage when tending a trap. The less food there is, 
the more advantages accrue from being sub-sized. On short ra- 
tions the genetically-small child has a better chance to live and 
come to maturity. In the extremes of heat, the small man is un- 
questionably favored, as is true also during violent exercise even 
at moderate temperatures. 

The small peoples of the world tend to be found nearer the 
equator, and there is a marked negative correlation between the 
mean annual temperature and weight. As one moves southward 
in Europe temperature rises and weight drops, as D. F. Roberts 
(1953) has demonstrated for 116 different populations the world 
around. The very lowest average weights (96-100 pounds) are 
associated with mean annual temperatures of 70-82° F., the 
highest average weights ( in excess of 160 pounds ) are associated 
with a mean average temperature of 40° F. (Fig. 12). 

As with any statistical association, the relevant variables are 
undoubtedly complex. Part is unquestionably physiological adap- 
tation. Russell Newman (1956) has demonstrated for America 
that the colder the state of origin, the more fat young men have! 
Part is in all probability genetic adaptation to extremes of heat 
and cold with the little men stemming from lands that are hotter. 











5 60 

5 20 

4 80 

4 40 


"■ - 

1 . 


s. ■■ 




© % x. 

• 9 # 

• og 







O IO 20 30 40 SO 60 70 80 

Mean Annual Temperature °F 

Fig. 12. Relationship between mean body weight and mean annual temper- 
ature. The higher the mean annual temperature, the lower the body weight 
of populations tends to be. (From Roberts, 1953, and Garn, 1960.) 

And part may be attributed to genetic adaptations to nutritional 
extremes in the overcrowded, "underdeveloped" equatorial lands. 
Famine is a powerful selective force, differentially eliminating 
the massive and large. Famine does not have to be a consistent 
part of the environment; one dies but once of starvation. In the 
face of continual caloric restriction, the genetically-small indi- 
viduals have a better chance to reach maturity and to reproduce 


Within each geographical race there is a range of sizes small to 
large. This range can best be comprehended in terms of the selec- 
tive forces, temperature for one, and the available food supply 
for another. However, these generalizations about body size are 
fully documentable but do not explain the Ithuri Forest Pygmies. 
Nor do they explain the surprising fatness of the Papago Indians 
of North America. These particular exceptions merit very particu- 
lar investigation and the metabolism of the Pygmies is currently 
being investigated by Dr. George Mann. 


Design a man for extreme cold and you have a virtual globule. 
Thick-set, reduced of leg and peripheres, there is a minimum of 
outstanding projections. By limiting surface relative to mass, less 
heat is loss by radiation, conduction and convection and (prob- 
ably) by insensible perspiration. By pulling-in projections closer 
to the warm body core, differential cooling is avoided with con- 
sequent danger of freezing. 

Design a man for dry-desert heat and he approaches paper- 
thinness. By maximizing surface relative to mass, heat loss by 
perspiration is maximized. Long hands and feet and a beaky bony 
nose, while contributing relatively little to the total cooling sur- 
face, are at no disadvantage where environmental temperatures 
are high relative to body heat. 

It is not quite appropriate to bring in a variety of species for 
comparison, selecting perhaps the rotund seagoing mammals at 
the one extreme, and the linear desert fox at the other. Neverthe- 
less, desert-adapted representatives of a species do tend toward 
linearity while related arctic forms exhibit a lower surface-mass 
ratio ( Bergmann's law ) . And, in man, there is a distinct tendency 
for the more linear groupings to be associated with high (and 
relatively dry) environmental temperatures, and for the thickest 
populations to inhabit areas where below-freezing temperatures 

Adaptations in body-build, however, involve more than just 
proportions, complicating the analysis of what we readily see and 
most easily measure. We must consider body composition, the 
amount of fat and especially the thickness of the insulating 



Fig- 13A 

Fig. 13A & B. Extreme differences in body proportions and the surface/mass 

ratio in Eskimos and in the Nuer. (Photographs, courtesy of the National 

Museet, Copenhagen and Evans-Pritchard, E.E.: The Nuer, Oxford, 

The Clarenden Press, 1940.) 



Fig. 13B 


blanket of outer fat. We must consider the location or "patterning" 
of the outer fat, which is subject to sexual as well as climatic 
selection. Fat over the cheekbones of Eskimos, even thin ones, 
and about their orbits, not only contributes to their flat-faced ap- 
pearance, but constitutes a protective mask as well. The stored fat 
on the rumps of Bushmen and Hottentots, an exaggeration of the 
typical hominid pattern, constitutes a particularly neat solution to 
two different problems. It provides a caloric reserve, an energy 
store for use during periods of food-scarcity, and being restricted 
in its location, least interferes with heat loss to the environment. 
Variations in body proportions, in the amount and patterning 
of fat, and in other bodily constituents (including the plasma 
volume and red-cell-volume ) are not the only cold-climate and 
hot- weather adaptations. As we have come to learn, each en- 
vironmental stress is met by multiple adaptations. But racial 
variations in body build are conspicuous and we are beginning to 
learn much about the "inner contours" of people of various races. 
Moreover, experimental studies can be devised to test their adap- 
tive values, to verify the directions that natural selection has taken 
in shaping the shapes of man. 


Of the climatic extremes into which man has ventured, none 
is more quickly lethal (and therefore more selective) than the 
extreme cold. Winter temperatures from minus 40° Fahrenheit to 
minus 90° F. are reported for various places in the inhabited 
Arctic. Without well-designed winter gear, such temperatures 
would be deadly in a matter of minutes. Even in full arctic regalia 
frostbite and death by freezing are ever present potentials, their 
severity unalloyed by the fact that temperatures within the snow- 
built igloo or sod barbara may be high. 

One obvious adaptation to cold-weather living is a minimum 
surface/mass ratio, with its obvious heat-conserving potentiality. 
A second involves size reduction in the limbs, particularly the 
legs, bringing them closer to the warm, central core. Both are 
characteristic of arctic peoples ( Fig. 13 ) . An increased thickness 
of subcutaneous fat would also be energy-conserving as Baker 
and Daniels (1955) have confirmed. However, Newman (1956), 


working with draftees, has indicated that such a response may 
be largely physiological rather than purely genetic. Of the vari- 
ous cold adaptations known in mammals, the ability to generate 
more heat by raising the metabolic rate would be applicable to 
man, provided of course, that enough food is available. Evidence 
from a number of experiments favors enhanced metabolic re- 
sponses to cold in Eskimos (Brown and Page, 1952), and in cer- 
tain other groups. 

Despite warm arctic clothing, the face is still exposed to wind 
and cold. The hands must be protected against freezing and must 
maintain their finer manipulative skills. Moreover, the peculiar 
Eskimo physiognomy (much more consistent across the Arctic 
than Eskimo serology) must also be noted in this connection. 
With fat-padded malars, fat-filled orbits, narrow eye-slits and re- 
duced nasal profiles, the Eskimo face fits the model of a cold- 
weather face. And a number of studies, like those of Brown and 
Page ( 1953 ) confirm higher peripheral skin temperatures during 
cold stress (Barnicot, 1959). By implication at least, manual 
dexterity in Eskimos would be less impaired at low environmental 

It should be emphasized that studies on cold adaptation have 
been so far limited by technical difficulties, subject selection and 
sample size. Laboratory experiments, moreover, fail to simulate 
natural selection where sheer survivorship rather than peripheral 
skin temperatures is the test of adaptive fitness. Nevertheless, the 
probability is that peoples long exposed to cold-selection will ex- 
hibit, on the average, features and mechanisms of value in the 


In contrast to the Arctic and its far below-zero temperatures, 
minimal night temperatures at near-freezing levels (or slightly 
below) may seem unimportant. But such frosty situations com- 
mand a very large part of the world, even the simmering deserts. 
And there are many human groups (quite recently some of our 
ancestors ) who, poorly-clad, were in danger of frosting their toes 
as they slept. The problem of peripheral skin temperatures at 
moderate levels of coldness further relates to American Negroes, 


who apparently experienced more cold injuries during the Korean 
conflict than other soldiers. 

Among the groups studied to date are the Alacaluf (the 
aborigines of Patagonia), desert-dwelling Australian aborigines 
and Bushmen of the Kalahari. All exhibited the ability to sleep 
nude or partly nude, often using their clothing as pillows under 
conditions quite uncomfortable for Europeans. Generally, they 
exhibited an enhanced ability to raise their metabolic levels and 
thus generate more heat except for the aboriginal Australians. 
Peripheral skin temperatures tended to be higher than in Euro- 
pean subjects under the same conditions (indicating greater 
peripheral blood flow). Again, the Australians' adaptations were 
in the reverse direction ( see Fig. 14 ) . 


A hot-humid environment is especially enervating, as we know 
in the sticky dog-days of summer. With moisture-saturated air, 
even below 100° F., the body becomes bathed in a continuous 
layer of sweat in an effort to reduce the heat load. In the tropics, 
there is attendant danger of excessive salt loss. If the body tem- 
perature rises too high, there is the possibility of death by circula- 
tory collapse. Even apart from this extreme direction of selection, 
hot-humid environments would tend to favor individuals who can 
maintain a moderate work load in the water-saturated atmosphere. 

For the existence of heat adaptations in man there is both direct 
and indirect evidence. The poorer cold-adaptation exhibited by 
American Colored subjects in a number of studies would suggest 
that they are relatively heat-adapted instead. Paul T. Baker 
( 1958 ) compared colored and white subjects, carefully matched 
and acclimatized, and concluded that the former were better 
adaped to humid heat ( but not to dry heat ) . 

Assuming that the long-term inhabitants of steamy climes are 
the product of selection by humid heat, what mechanisms are at 
least theoretically involved? Dark skin color, by raising the sur- 
face temperature, could bring about sweating earlier. An in- 
creased sweating rate involving more sweat glands has been sug- 
gested, but is currently the subject of debate as summarized by 
Barnicot (1959). A lower rate of salt loss is among the other 

| 111 1 1 1 1 

01 ~ 

• -*v 










\ r 







Q-, c 


E I5- 




\ A 

•• \ 




CO o 


1 1 1 1 1 I i i 
Dl 2345678 


Fig. 14. Adaptation to moderate cold during sleep. In desert-dwelling 
Australian aborigines adaptation to near freezing night temperatures involves 
a decrease in peripheral skin temperature. The marked elevation of the 
relative metabolic rate shown in control whites is not exhibited by the 
aborigines. (From Hammel et al., 1959.) 


mechanisms currently being studied. A fourth possibility — known 
for some desert animals — is the ability to tolerate increased heat 

Logically, the peoples of the African rain-forest and those of 
the Amazon basin exhibit adaptations to humid heat, though not 
necessarily the same adaptations. For both groups, of particular 
theoretical interest, the direct evidence is at present minimal. 


Desert living is essentially a compromise, requiring tolerance of 
mid-day heat and night cold, tolerance of high ultraviolet intensi- 
ties without increasing the heat load, and the ability to lose heat 
by perspiration while conserving precious water. 

Desert mammals suggest several pathways to arid living, includ- 
ing nocturnal habits, leanness and small body size. Only the latter 
two appear to be applicable to man, who notably ventures into the 
noonday sun. The early loss of subcutaneous fat in the Bushman 
may indicate its inadaptive nature under Kalahari conditions and 
the Bushmen are notably small, as are many other desert peoples. 

In theory, skin pigmentation should be moderate in desert men, 
enough to protect the malpighian layer, but not enough to build 
up the heat load. Baker ( 1958 ) has shown the disadvantages of 
extremely dark skin under hot-dry conditions. Thus the lighter 
pigmentation of the Bushmen may represent the theoretical com- 
promise situation, a balance between competing directions of 

A third area of desert specialization particularly exhibited by 
the Kangaroo rat Dipodomys is the increased ability to concen- 
trate urine. If man could follow the lead of this desert-living rat, 
he could conserve one to two pints of water a day, not a great 
deal compared to water lost in the exhaled air, but a saving none- 
theless. No evidence exists at present for or against the possi- 
bility of increased urine concentration in desert man, a possibility 
further complicated by their largely protein dietary. However, 
there are two disorders in which urine is less concentrated than 
usual: the first is fibrocystic disease of the pancreas. Heterozy- 
gotes for this disease, therefore, would be less viable under desert 





Hammel et al. 

Scholander et al. 

Wyndham and 
Morrison (1958) 


8 Central Australian 

6 Control "whites" 

9 Tropical Australian 

5 Australian whites 

6 Central Australian 


Small but important differences 
in thermal and metabolic re- 
sponses to moderate cold during 

Europeans maintained body heat 
by increased muscular movement 
during sleep. 

2 Unacculturated Bushmen No important differences found 

2 Europeans in response to moderate night 


Hammel (1960) 

9 Alacaluf 

Markedly increase metabolic re- 
sponse to night cold, persistent 
elevated BMR. 

Adams and Covino 
(1958) (see also 
Brown and Page 

Lampietro et al. 

Bass et al. (1959) 

Baker (1958) 

Wyndham et al. 

7 Negro soldiers 

7 "Caucasian" soldiers 

6 Anaktuuk Eskimos 

17 white and 

16 Negro volunteers 
matched for body size 
and composition 

16 East Indians 

16 U. S. Negroes 
23 Chinese 

17 U. S. Whites 
8 Eskimos 

Systematic Negro-white Eskimo 
differences in skin temperature 
and BMR during 2-hour exposure 
to cold. 

Fewer rewarming cycles and 
lower finger temperatures in 
Negro subjects exposed to 
moderate cold. 

Marked racial differences in 
plasma volume and blood volume 
especially for Eskimo of ques- 
tionable relationship to climatic 

40 pairs of white and Negroes displayed a higher physi- 

Negro soldiers matched ological tolerance to hot humid 

for body composition and conditions, but heated up more 

size under hot dry conditions. 

8 African mine laborers 
White data from the 

Lower sweating rates, lower heart 
rates and lower rectal tempera- 
tures for Africans 

See also Barnicot 


conditions, and it is a safe bet that the gene involved will prove 
rarer in the hotter, dryer parts of the world. 

The second gene-determined condition known to be unfavor- 
able in the desert is the sickling trait ( see chapter VI ) . Peculiarly 
enough, the heterozygote NS has a reduced ability to concentrate 
urine, and is therefore at an adaptive disadvantage where water 
is scarce ( Keitel, et at., 1956 ) . Needless to say, the sickling trait 
uncommon in the less watered parts of Africa and Saudi Arabia, 
in part because the heterozygote is at no advantage in such areas, 
and in part because the trait is disadvantageous where man's 
water loss is especially critical. 


Increasingly, population differences in body size, body-build 
and proportions and skin pigmentation may be viewed as adapta- 
tions to particular climatic extremes. Such differences often trans- 
cend geographical races while the distribution of the differences 
provide a hint as to the origin of particular races. 

However, a particular climatic stress will not always be met 
by the same adaptation. Therefore, the adaptive nature of racial 
differences must be determined under conditions simulating the 
environments in which they have arisen, and using acclimatized 


* Adams, T. and Covino, B. G.: Racial variations to a standardized cold 
stress, /. Appl. Physiol, 12:9-12, 1958. 

Baker, P. T.: American Negro-White Differences in Heat Tolerance, 
H.Q. Quartermaster Res. and Engin. Command, U. S. Army Techn. 
Dep. EP-75, 1958. 

Baker, P. T.: Racial Differences in heat tolerance, Am. J. Phys. An- 
thropol, N.S. i6:287-305, 1958. 

Baker, P. T. and Daniels, F. Jr.: Relationship between skinfold thick- 
ness and body cooling for two hours at 15° C, /. Appl. Physiol., 
8: 409-416, 1955. 

Barnicot, N. A.: Climatic factors in the evolution of human popula- 
tions, Cold Spring Harbor Symposia on Quantitative Biology, 24: 
115-129, Cold Spring Harbor, The Biological Laboratory, 1959. 


Bass, D. E., Iampietro, P. F. and Buskirk, E. R.: Comparison of basal 

plasma and blood volumes of Negro and white males, /. Appl. 

Physiol, 14:801-803, 1959. 
Blum, H. F. : Ultra-violet radiation and cancer, in Hollaender, A. ( ed. ) : 

Radiation Biology, Washington, National Academy of Sciences, vol. 

II, 529-559, 1955. ' 

Brown, G. M., Hatcher, J. D. and Page, J.: Temperature and blood flow 

in the forearm of the Eskimo, /. Appl. Physiol, 5:410-420, 1953. 

* Brown, G. M. and Page, J.: The effect of chronic exposure to cold on 

temperature and blood flow of the hand, /. Appl Physiol, 5: 221-227, 

Ellinger, F.: Medical Radiation Biology, Springfield, Thomas, 1957. 
Hammel, H. T.: Response to cold by the Alacaluf Indians, Current 

Anthropol, 1:146, 1960. 
Hammel, H. T., Eisner, R. W., Le Messurier, D. M., Andersen, H. T. 

and Milan, F. A. : Thermal and metabolic responses of the Australian 

aborigine exposed to moderate cold in summer, /. Appl Physiol, 

14:605-615, 1959. 
Iampietro, P. F., Goldman, R. F., Buskirk, E. R. and Bass, D. E.: 

Response of Negro and white males to cold, /. Appl Physiol, 14: 

798-800, 1959. 
Keitel, H. G., Thompson, D. and Itano, H. A.: Hyposthenuria in sickle 

cell, anemia: a reversible renal defect, /. Clin. Invest., 35:998-1007, 

Newman, R. W.: Skinfold measurements in young American males, 

Human Biol, 28:154-164, 1956. 
Newman, R. W. and Munro, E. H.: The relation of climate and body 

size in U. S. males, Am. J. Phys. Anthropol, N.S. 13:1-17, 1955. 

* Roberts, D. F.: Body weight, race and climate, Am. J. Phys. 

Anthropol, N.S. 11 .533-558, 1953. 

Scholander, P. F., Hammel, H. T., Hart, J. S., LeMessurier, D. H. and 
Steen, J.: Cold Adaptation in Australian aborigines, J. Appl Physiol, 
13:211-218, 1958. 

Schreider, E.: Geographical distribution of the body weight/body 
surface ratio, Nature, 165:286, 1950. 

Wyndham, C. H., Bouwer, W. v. d. M., Devine, M. G. and Paterson, 
H. E.: Physiological responses of African laborers at various satur- 
ated air temperatures, wind velocities and rates of energy expendi- 
ture, /. Appl. Physiol, 5:290-298, 1952. 

Wyndham, C. H. and Morrison, J. F.: Adjustment to cold of Bushmen 
in the Kalahari desert, /. Appl. Physiol, 13:219-225, 1958. 



JTor generations mankind has lived with malaria, or 
to be more exact, has lived despite malaria. Great sections of the 
world have proved unhealthy, malarious, poor places to survive 
In both hemispheres, countries situated between 35° N and 20° S 
have been cradles of malaria, the "shaking ague." In Europe and 
the Middle East malaria had control of both sides of the Adriatic, 
much of the Italian coastline, Greece, the Ionian Islands and 
Crete, and the shores of the Black Sea and the Caspian Sea. Since 
the malarial parts of the world, including the New World, were 
generally well-watered and usually densely-populated areas, it is 
not surprising that in 1880 one half of the entire mortality of the 
human race was attributed to malaria! 

In the areas of endemic malaria, in southern Europe and East 
Africa, it was observed that some few individuals did not acquire 
malaria, or at least did not exhibit the chills and recurrent high 
fevers, the enlargement of the spleen and intestinal symptoms 
characteristic of the disease. Various explanations were offered, 
especially in the days when malaria was blamed on night air and 
exhalations of the marshes. Still, no really satisfactory evidence 
was offered as to whether immunity to malaria was natural or 
acquired early, whether it was due to some chance event ( such as 
being unattractive to mosquitoes ) or whether malarial immunity 
existed at all. 

Rather recently, and as a by-product of studies on hereditary 
blood disorders, the existence of true natural immunity to malaria 
has been confirmed. Even more important, the relevant mecha- 
nisms have been discovered, and the disease of malaria has been 
shown to be responsible for several important directions of poly- 
morphism in man. 





The story begins in the Mediterranean, in Italy and Sicily, 
Greece, and some parts of the Middle East. There, the disease 
called thalassemia is common, a disease that has two distinct 
forms, thalassemia major and thalassemia minor. 

Certain interesting features of thalassemia soon made them- 
selves evident. The first, of course, was the Mediterranean distri- 
bution of the disease, especially (as noted in the United States) 
among individuals of Italian or Greek ancestry, but also among 
North Africans, Egyptians and peoples from Asia Minor. Though 
observed elsewhere in the world (for example, Thailand) but 
rarely among Europeans not from the Mediterranean area, thalas- 
semia is for practical purposes primarily limited to the Mediter- 
ranean and Irano-Mediterranean local races of the European 
Geographical Race ( see chapter XI ) , and to local races in India. 




Clinical picture 





Within normal 

Normal, slight 


limits for hemoglobin 

amount of 

and cellular fragility 

fetal hemoglobin 



Slight anemia, 

slight amount 


increased osmotic 
pressure of the red cells 

of fetal hemoglobin 



Marked anemia, 



abnormal red cells 

primarily of fetal 

fragile and increased 

type, little normal 

osmotic pressure 


From Ingram and Stretton (1959), Neel (1949, 1950). 

Family-line studies show that thalassemia is hereditary and that 
parents of a person with thalassemia major are "carriers" of the 
disease and exhibit thalassemia minor. Thalassemia is inherited 
as a dominant with thalassemia minor the heterozygotic state and 
thalassemia major the homozygotic. From the observed incidence 
of the disorder it is possible to calculate the gene frequency which 
ranges from nearly zero in Switzerland and northward in Europe 


to 0.20 or more in certain areas of Cyprus and northern Italy. For 
much of Italy, where epidemiological data is quite complete, 
frequencies for thalassemia average under 0.02, from 0.02 to 0.06 
in Sicily and Corsica, 0.10 in the coastal area northeast of Bologna 
and 0.18 in the Ferrara region. 


Epidemiological and family-line data together confirm the 
existence of thalassemia as simply inherited in the form of a Men- 
delian dominant and with gene frequencies exceeding 0.02 and 
sporadically as high as 0.20 in the Mediterranean coastal area. 
Biochemical studies on the blood of individuals suffering from 
thalassemia major (the homozygotes) and other studies on the 
heterozygotic individuals suggest that the defect is primarily in 
the ability to produce normal adult hemoglobin A. Thus, the 
blood of sufferers from severe Cooley's anemia is almost entirely 
of the fetal type ( or possibly of the A 2 hemoglobin subtype ) while 
in thalassemia minor, the proportion of fetal hemoglobin is low. 
In short, the homozygotes cannot produce adequate amounts of 
the normal adult hemoglobin A, and their symptoms are largely 
referable to this gene-determined hemoglobin deficiency. 

It must be noted that the individuals homozygous for the 
abnormal hemoglobin, that is sufferers from thalassemia major, 
rarely reach reproductive age. In fact, Neel (1950) doubts 
whether the homozygotes could reproduce even if they lived to 
the child-bearing period. This lethality obviously eliminates some 
proportion of the thalassemia genes in every generation. 


The clinical and geographical data on thalassemia raised two 
important problems, one being the origin of the abnormal gene 
( and its seeming restriction to the Mediterranean ) and the other 
being its continuance in time. From the wide distribution of the 
gene on both shores of the Mediterranean, an ancient origin could 
be postulated — prior to the spread of the Mediterranean local race 
as early as 5000 B.C. However, the absence of thalassemia in 
parts of Europe where Mediterraneans have migrated remained 
a puzzling feature. Even more peculiar was the continuation of 


the abnormal gene. Since, in each generation, a certain propor- 
tion of the abnormal genes are lost, the mutation would ordinarily 
be expected to remain at a low level — balanced only by new 
mutant genes. 

What then, explains the continuation of, the abnormal gene, and 
its exceptionally high incidence in Crete, Bologna and Corsica? 
One explanation would be an increased fertility in heterozygotes, 
but such increased fertility has not been found and could not 
explain known "islands" of thalassemia. Similarly, an abnormally 
high mutation rate for the thalassemic gene, though postulated, 
has not been found, and again would not explain the marked 
variations in the incidence of thalassemia in Italy, Greece and 
elsewhere from Iran to South China. 

Beginning in 1950, however, a number of workers pointed to 
the correspondence between the distribution of thalassemia and 
the incidence of malaria. In those parts of Europe where malaria 
was holoendemic, that is a severe problem the year round, thalas- 
semia frequencies were highest. In the cold-climate and higher 
areas, thalassemia frequencies drop to near zero. Working in 
Sardinia, Cepellini (1955) has shown a far higher incidence of 
thalassemia in low-lying regions where malaria is holoendemic 
than in corresponding villages at higher altitudes. The emerging 
picture, and one now well supported by findings on the sickle-cell 
disease which follows, is that the thalassemic gene affords protec- 
tion against malaria. 

In a malarial region the homozygotes for thalassemia die early 
without reproducing. The "normal"' homozygotes are afflicted 
with malaria and frequently die early, while the heterozygotes are 
somewhat protected and thus the continuance of both genes is 
assured. This simple explanation fits the data, explains the non- 
uniform distribution of thalassemia in the Mediterranean and 
makes the race-limited nature of thalassemia comprehensible in 
terms of natural selection in malarial areas. 


A second hereditary blood disorder, primarily African in its 
distribution, is called sickle-cell anemia, after the characteristic 
appearance of the red blood cells in saline solution. As with 


thalassemia, sickling is inherited as a Mendelian dominant and 
there are two forms — the mild sickle-cell trait ( the heterozygotic 
state) and the severe sickle-cell disease (the homozygous state). 

The inheritance of sickling was confirmed by Neel in 1949 who 
demonstrated that individuals with sickle-cell disease were in- 
variably the offspring of two parents with the sickle-cell trait. 
The molecular nature of sickling, in turn, was reported by Linus 
Pauling in the same year and the existence of a number of ab- 
normal hemoglobins was subsequently discovered by Neel and his 
associates. Briefly, the individual homozygous for the sickle-cell 
trait produces the abnormal hemoglobin S (or some other ab- 
normal hemoglobin) which has less oxygen-containing ability 
than the normal hemoglobin A. In a typical East African popula- 
tion, as in the American Colored population, there are three 
genotypes — the homozygous "normals," the heterozygous individ- 
uals with the sickle-cell trait and those homozygous for the ab- 
normal hemoglobin, exhibiting the sickle-cell disease (Fig. 15.). 

In Africa, frequencies of the sickling trait vary widely from 
zero in some areas to as high as 40% in other areas. Initially, these 
marked variations were explained in area-incidence terms ( assum- 
ing the origin of the gene in the areas of highest concentrations ) 
or by postulating migrations from high-sickling areas to low- 
sickling areas. Such explanations, reminiscent of racial anthro- 
pology of the last century, and early explanations given for 
variations in the blood-group frequencies, put maximum weight 
on hypothetical migrations and minimum weighting on purely 
local, epidemiological factors. 

However, as with thalassemia, there had to be a more dynamic 
explanation for both the continuance and the peculiar distribution 
of the sickle-cell trait than migration theory offered. Since the 
sickle-cell disease is usually lethal, and since there is a differential 
elimination of individuals with the sickle-cell trait (cf. Raper, 
1949), some selective advantage had to maintain the sickling 
gene. A number of workers, among them Allison (1954) noted 
the geographical association between malaria and sickling (Fig. 
16. ) and postulated an adaptive advantage for the heterozygote 
in malarial areas. This postulate has now been well documented 
in a number of ways as reviewed by Allison in 1955. 



nkn1 ! i] i both sickling and 

normal genes. 
nVs! Ms] [nYs 








Fig. 15. The mechanism of "sickling" and the differential survival of indi- 
viduals with the sickling trait where malaria is a severe problem. 







LakeVictoriaWXf q i 








Fig. 16. Association between the frequency of the sickle-cell trait in East 
Africa (designated by the percentages drawn on the map) and the preva- 
lence of malaria (as shown by stippling). Where malaria is hyperendemic 
the percentage of sicklers is highest. (Redrawn from Allison [1955] and 
reprinted from Readings on Race, p. 200.) 

By now, the selection advantage attached to the sickling heter- 
ozygote has been confirmed, particularly by investigating children 
in African areas of hyperendemic malaria. In such areas, the 
homozygous normals develop malaria early, many die, and the 
vitality of the survivors is impaired. Those individuals homo- 
zygous for the sickle-cell trait develop sickle-cell disease with 
consequently increased mortality. The heterozygous sicklers, 
however, those with the sickle-cell trait but not the disease, are 
in effect protected. The incidence of malarial infestation is less. 
Heterozygotes develop malaria— but to a milder degree and with 
fewer effects particularly during pregnancy (Allison, 1955). 

Thus, sickle-cell disease in Africa, like thalassemia in the Medi- 
terranean (and in Greece where both are found) is an example of 
adaptive polymorphism in man. Both the normal gene, which 
serves to elaborate hemoglobin A, and the abnormal gene (which 


is responsible for hemoglobins S, etc. ) continue to exist in Africa 
because the heterozygote is at an adaptive advantage. The more 
common malaria is, the higher the gene frequency for sickling, 
reaching toward 0.50 as a maximum. Bu{ in areas where there is 
no malaria, or where malaria has come under control (as in 
western Europe and the United States ) the heterozygote is at no 
advantage. Thus DDT, marsh drainage, screens, mosquito re- 
pellents and antimalarials will eventually reduce the incidence of 
the sickling gene. 

It will be noted that the sickle-cell trait in Africa transcends 
the conventional boundaries of African local races. Even within 
related populations, the incidence of sickling may be far higher 
in one group than in another. To some extent, these local differ- 
ences may be due to the presence of other abnormal hemoglobins 
( C, E, etc. ) . But the greater part of the diversity is more easily 
attributed to purely local factors, chief among them the preva- 
lence of malaria. Confirmatory evidence comes from Saudi Arabia 
where the incidence of sickling is far higher among oasis-village 
dwellers than among the hill tribes from whence these same 
villagers stemmed. All of this adds to the mounting evidence that 
the genetic makeup of any population strongly reflects the con- 
ditions under which it lives. 


While the relationship between malaria and hemoglobins is 
quite clear in Africa, the existence of malaria itself poses a certain 
problem. Much of Africa is not naturally malaria territory: in its 
pristine state Africa offered few places for the mosquito Anoph- 
eles gambiae to breed. Only where the rain forest has been 
opened up by primitive agricultural practices (or in modern 
rubber plantations ) does A. gambiae have a chance to spread. In 
the rain forest, or on the shaded forest floor, the malaria-carrying 
mosquito does not breed, and in the untouched rain forest malaria 
does not exist. 

Reviewing such information, Frank B. Livingstone (1958) has 
concluded that malaria is of recent introduction in West Africa, 
following the spread of slash-and-burn agriculture, the opening of 
the forest floor and the appearance of stagnant but not shaded 


pools for the carrier mosquito to breed in. Man made Africa 
malarial by providing the kind of climate the Anopheles gambiae 
mosquito needs, and by raising the human population density to 
the point where there were always new individuals to infect and 
thus spread the chain of malaria. The dissemination of malaria in 
Africa would thus follow the spread of slash-and-burn agriculture 
and the abnormal hemoglobin-S would thus become adaptive in 
its wake. Sickling frequencies, particularly in West Africa, ap- 
parently bear this hypothesis out; peoples still pre-agricultural or 
late in attaining agriculture have the lowest incidence of sickling. 


Livingstone has suggested that man paved the way for malaria 
in Africa; by cutting clearings in the jungle; man gave A. gambiae 
a place to breed and the resulting spread of malaria put an adap- 
tive premium on the otherwise inadaptive hemoglobin-S. This 
would appear to be the first concrete example of the way culture, 
that is learned behavior, is capable of bringing about genetic 
change within populations. 

The man-mosquito-malaria relationship in West Africa in- 
evitably suggests a review of man and malaria in the Mediter- 
ranean. Seemingly, Sicily, Sardinia, the Po valley and Cyprus 
have no immediate parallel with Africa. There were no jungles 
to cut down and while there has been deforestation in some areas, 
swamps such as the Pontine marshes can hardly be considered as 
recent human artifacts except by silting-up of rivers. 

Nevertheless, while man did not create the wet, humid marshy 
lands of the Mediterranean, he certainly did move into them for 
the practice of lowland agriculture. The early civilizations of the 
ancient world grew up along rivers with their seasonal floods and 
consequent stagnant pools. Oasis villages (where malaria is holo- 
endemic even today) provided the basis for early stable popula- 
tions. Primitive ditch-irrigation, wheel-and-bucket watering and 
crops specifically adapted to seasonal flooding supported the 
bulk of the populations. The consequent food surpluses make 
large population densities possible, and vastly increase the num- 
bers of people subjected to malarial selection. 



In the Mediterranean area through to Iran, and in a broad 
area of Africa, there are two distinct hereditary blood disorders 
that are irrevocably linked to malaria. , 

Thalassemia (or Cooley's anemia) is a hereditary disease, a 
minor disadvantage in the heterozygote ( thalassemia minor ) but 
lethal in the homozygote (thalassemia major). The rather high 
gene frequency for thalassemia in the southern Mediterranean 
appears to be maintained by malaria, generation after generation. 

The sickle-cell trait (involving the abnormal hemoglobin-S ) 
similarly reflects adaptive polymorphism. In malarial areas of 
Africa the normal homozygote and the abnormal heterozygotes 
are both at a disadvantage; the heterozygote has peak adaptive 
fitness where malaria is holoendemic. 

In all likelihood the spread of malaria in Africa, and therefore 
the abnormal hemoglobins, was brought about by man's opening 
the forest roof by slash-and-burn agriculture. In the Mediter- 
ranean, lowland agriculture may have maximized population ex- 
posure to malarial selection. In both regions changes in the genet- 
ical makeup of populations can clearly be attributed to particular 
cultural practices. 


Allison, A. C: Protection afforded by sickle cell trait against subtertian 

malarial infection, Brit. Med. J., 1:290-292, 1954. 
Allison, A. C: The distribution of the sickle-cell trait in East Africa 

and elsewhere, and its apparent relationship to the incidence of 

subtertian malaria, Tr. Roy. Soc. Trop. Med. and Hygiene, 48:312- 


* Allison, A. C: Aspects of polymorphism in man, Cold Spring Harbor 

Symposia on Quantitative Biology, 20:239-255, Cold Spring Harbor, 

The Biological Laboratory, 1955. 
Ceppellini, R. : The usefulness of blood factors in racial anthropology, 

Am. J. Phys. Anthropol, N.S. 13:389 (abstract) 1955. 
Ingram, V. M. and Stretton, A. O. W.: Genetic basis of the thalas- 

saemia diseases, Nature, 184:1903, 1909, 1959. 

* Livingstone, F. B.: Anthropological implications of sickle-cell gene 

distribution in West Africa, Am. Anthropol., 60:533-562, 1958. 



Neel, J. V.: The inheritance of sickle-cell anemia, Science, 110:64-66, 

Neel, J. V. : The population genetics of two inherited blood dyscrasias 

in man. Cold Spring Harbor Symposia on Quantitative Biology, 15: 

141-158, Cold Spring Harbor, The Biological Laboratory, 1950. 
Pauling, L., Itano, H. A., Singer, S. J. and Wells, I. C: Sickle-cell 

anemia, a molecular disease, Science, IM-543-548, 1949. 
Raper, A. B.: Sudden death in sickle-cell disease, E.Afr. Med. J., 26: 

14-22, 1949. 
Raper, A. B.: Malaria and the sickling trait, Brit. Med. J., 1.1186-1189, 

* Reprinted in Readings on Race. 



Oince races are natural units, reproductively isolated 
from each other and with separate evolutionary histories through 
time, it is not surprising that they differ from each other in a great 
many gene-determined respects. Considering the unique history 
behind each race, and the geographical and ecological uniqueness 
of its successive homelands, lack of differentiation would be re- 
markable indeed. Particularly in the random loss or chance ac- 
quisition of genes, each race represents a cumulative succession of 
accidents that could never be duplicated in millions of years. 

These statements about racial differentiation, applicable to the 
"normal" genes commonly considered, pertain to abnormal or 
disease genes equally well. For the rare diseases, represented by 
but few carriers even in large populations, chance events could be 
prepotent. At a gene frequency of 0.01, 0.001 or even less, chance 
might easily eliminate a disease gene in one population while 
doubling or even tripling its frequency in a second. In one breed- 
ing population, the requisite disease mutation might never have 
happened, while in yet another population that mutation could 
have taken the form of a run. 

Checking upon gene frequencies for rare diseases is at present 
no easy task. In most areas of the world, lacking hospital facilities 
and diagnostic skills, population comparisons are impossible. 
Even in the great medical centers, rare hereditary disorders may 
easily be missed, while other disorders may erroneously be con- 
sidered race-limited or population-limited in their stead. Obvi- 
ously, even elementary data on the comparative frequencies of the 
rare gene-determined disorders are to a large extent lacking today. 

Nevertheless, there are numerous genetic disorders, either rare 
or lacking in most populations, that reach major proportion in a 



few groups. The sickle-cell trait, mentioned in the previous chap 
ter with its counterpart sickle-cell anemia, is one example, as is 
Mediterranean anemia (thalassemia major). For these gene- 
determined diseases, with frequencies close to zero in most parts 
of the world, trait frequencies reach 0.2 to 0.4 in a few restricted 
areas. In both examples the abnormal gene is maintained because 
the heterozygote is at an adaptive advantage. 

A second class of disorders almost unknown in most parts of 
the world, but uniquely common in one restricted area, is in- 
creasingly coming to our attention. One of these unique diseases 
is Kuru, an extraordinary hereditary neurological disorder appar- 
ently restricted to Eastern New Guinea. A similar apparently 
population-limited neuromuscular heredofamilial disease has been 
reported from the Trust territory. In the Mediterranean area, 
familial Mediterranean fever and Favism are not only population- 
limited but promise to become included, along with the abnormal 
hemoglobins, among the conditions clearly adaptive to the hetero- 
zygotic state. 

The third category, that of genetic diseases rare even in the 
populations that have them is beginning to yield information. 
Leprechaunism appears to be an "Irish" disease, in that all known 
cases are of Irish origin or descent. Tay-Sachs disease and familial 
dysautonomia are probably "Jewish" diseases. The most economi- 
cal explanation, that of genetic drift, probably explains these 
population-limited disorders. However, there are beginning sus- 
picions that natural selection is also at work. By way of example, 
differences between the Ashkenazic ( European ) Jews and the 
Sephardic (Spanish-Portuguese and North African) Jews sug- 
gest that local environmental factors are important even with 
respect to the really rare hereditary diseases. One can easily 
understand how a disorder, such as congenital ectodermal dys- 
plasia, where sweat glands are largely lacking would be more 
likely to survive in Stockholm than in Salerno. 

Mutations being what they are, no disease can be considered 
to be entirely race-limited. The same abnormal hemoglobin that 
protects Italians in malarial areas also exists in Burma. The gene 
for familial dysautonomia probably also exists in some Moslem, 
Shinto and Buddhist groups but so far the disease has been re- 


ported only among Jews. Nor should similarities in the occur- 
rence of rare hereditary disorders necessarily indicate common 
ancestry. The distribution of the sickling gene in Africa, Mada- 
gascar, Yemen and Saudi Arabia we know now to be influenced 
by environmental selection. Thus, similarities between races in 
the same ecological zone, traditionally attributed to admixture, 
are often due to common directions of selection. 


Our first example is Kuru, a remarkable disease apparently 
limited to the Eastern Highlands of New Guinea. Unknown to 
medical science until 1953, and quite unstudied until 1957, Kuru 
is well described by its pidgin-English name "skin-Guria" meaning 
shaking. Kuru is a progressive and incurable neurological dis- 
order. Within a year after the onset of symptoms, the afflicted 
individual ordinarily dies. 

Typically, the first sign of Kuru is incoordination. The victim 
begins to stumble, then increasingly he becomes less able to walk 
and involuntary tremors become more and more common. Soon 
he is no longer able to sit, and speech becomes unintelligible. 
Next the abilities to swallow food, to urinate and to defecate are 
no longer under the individual's control. Commonly, the full 
course of Kuru is run in a year or less, but the complete progress 
of this disease may take as little as three months. 

Peculiarly, Kuru is limited to one group in Eastern New Guinea, 
the Fore, and some neighboring people among whom Fore women 
have married. It is known that far more women than men are 
afflicted by Kuru so that the female/male ratio among affected 
individuals is 14-1 or even higher through the third decade of life. 
Possibly 1% of all Fore natives are suffering from Kuru at any 
time, and in some hamlets as much as 50% of all deaths are due to 

In an effort to find a trace element, poison, or food deficiency 
responsible for the diseases, Fore food, Fore body paints and 
even Fore campfire smoke were tested. No trace element or rare 
earth that could serve as a nervous-system poison has been found. 
Moreover, Fore men living on Government dietaries outside of the 
Fore area also develop Kuru, so that Kuru is not a nutritional- 


deficiency disease. All evidence points clearly to Fore ancestry, 
and not the environment in which the Fore live. On one side of 
the formidable Lamari river the Fore have Kuru, but on the other 
side the Kukukuku people who have not intermarried with the 
Fore do not have Kuru ( Figure 17 ) . To the south, the Yar do not 
suffer from Kuru, but Fore who have ventured through the forests 
into the isolated Yar territory do develop Kuru. Fore women who 
marry into the Keiagana, Kanite and Kimi tribes develop Kuru 
but their hosts do not, thus proving that the disease is not com- 

Kuru appears to be an hereditary disease, inherited as a Men- 
delian dominant, but hormone-mediated. The gene frequency is 
estimated as approximately 0.37. In the female the homozygote 
develops early Kuru whereas the heterozygote develops late 
Kuru. Among the males the homozygotes die of early Kuru while 
the heterozygotes survive as do the homozygous normals. Since 
most of the heterozygous females live through the reproductive 
period and even those homozygous for Kuru (married early in 
life) manage to have children, the continuation of the abnormal 
Kuru gene is therefore assured. 

Still, one may ask what keeps up the rather high gene fre- 
quency for Kuru. After all, in each generation, many Kuru genes 
are removed from the population. Even the progeny of individ- 
uals heterozygous for Kuru are relatively fewer. Since an ab- 
normally high mutation rate for Kuru is a most unlikely possi- 
bility, some adaptive advantage for the heterozygotes may well 
be sought. 

Carle ton Gajdusek, the outstanding American authority on 
Kuru, observes that "leprosy and yaws are less frequent here 
(in the Fore) than in many surrounding populaces who do not 
suffer from Kuru." Obviously, if the Kuru-gene protects against 
either disease, it could counteract the loss of genes due to Kuru. 
Another possibility is that childhood mortality is lower in hetero- 
zygous individuals, although the statistics to date are not impres- 
sive. Still, a differential survival of heterozygous males in early 
childhood could help to balance the loss of Kuru genes in later 

But the Fore have an interesting custom in regard to Kuru, a 











Fig. 17. The Fore territory in eastern New Guinea where Kuru is common. 
On the east side of the Lamari River Kuru is not found, whereas the gene 
frequency is as high as 0.38 among the Fore. (Redrawn from Gajdusek 

and Zigas, 1959). 


custom that may explain the continuance of the gene. The Fore 
think that Kuru is the result of sorcery. When someone dies of 
Kuru, the Fore people seek out, attack and usually kill the sus- 
pected sorcerer. This custom, called tukabu would have impor- 
tant genetic effects. It would remove more normal genes than 
Kuru genes since the ratio of normal to Kuru genes in the popula- 
tion as a whole is 6:4. However, in adulthood the disparity is even 
greater, with a continued removal of Kuru-prone homozygous 
individuals. Thus, in adults, two or three normal genes might be 
removed from the population for each abnormal gene lost to 
Kuru. It may well be, then, that tukabu is a partial answer to the 
continuance of Kuru among the Fore. Meanwhile the Territorial 
Government has quarantined the entire Fore tribe.* 


The borders of the blue Mediterranean overlook a long coast- 
line, largely inhabited by a single local race. There, over great 
antiquity, the great civilizations and the great religions of the 
Western World arose. And in the Mediterranean there appeared 
a rare hereditary disorder that deservedly bears the name, familial 
Mediterranean fever. 

Mediterranean fever is a "periodic" disease. Once the symp- 
toms, the fever and malaise have begun, they recur sporadically 
and unpredictably during the individual's lifetime. At the least 
there is fever, lasting a day or two, joint pains and chest and 
abdominal pain. In advanced cases, there is joint involvement, 
decalcification of the bone and kidney insufficiency. Though most 
of the affected individuals are not permanently or seriously dis- 
abled, about 10% of cases studied to date succumbed to renal 

The familial nature of Mediterranean fever is clear-cut. About 
half of the siblings of index cases and about 50% of the offspring 
of index cases develop Mediterranean fever, thus suggesting a 
Mendelian dominant. There is a slight excess of males over fe- 
males among the known victims but an autosomal gene is clearly 

So far a total of 249 cases of Mediterranean fever have been 

* Science, 132:11, I960: South Pacific Post, May 24, 1960. 


described in the medical literature. Of this total, at least 235 
definitely stemmed from countries on or near the Mediterranean. 
The incidence appears to be particularly high among Armenians, 
Jews, Lebanese and Syrians as shown in the table below. 


Stated Origin Number of Patients 

North America 67 

Libya 25 

Egypt 5 

Israel 2 

Syria 1 

Lebanon 24 

Iraq 22 

Armenia 58 

Cavasus 4 

Turkey 14 

Bulgaria 2 

Greece 3 

Italy 3 

Spain 5 

Not stated 18 

* From Heller, Schar and Sherf ('58). 

Interestingly, among 73 Jews with this disease studied in Tel 
Aviv, not a single one was Ashkenazic (the group including most 
European Jews). Heller, Schar and Sherf ('58) also observed 
that Sephardic (i.e., North African and Oriental Jews) contrib- 
uted all of the Jewish cases described by other authors. 

The facts are not incompatible with the idea that the mutation 
for familial Mediterranean fever is an ancient one in the Medi- 
terranean as judged by its present wide dispersion. The mutation 
could well be 5000 to 6000 years old. The apparent concentration 
of the gene in the Eastern Mediterranean could be an indication 
of the site of origin, using the age-area approach, or it could be 
due to differential selection in the Bible lands. 

Since the gene frequency of familial Mediterranean fever is 
low, well under 0.0001, the present limits may well be due to 
chance. However, since familial Mediterranean fever is definitely 
associated with impaired fertility, and a certain proportion of the 
genes are removed from the population in each generation, there 


is need to explain how the gene frequency is maintained. The fact 
that Ashkenazic Jews do not exhibit the disease, while Sephardic 
Jews do, further requires explanation. A very likely possibility is 
that the heterozygote is at an adaptive advantage within the 
Mediterranean region, but not in northern, colder climes. 


Further information on racial differences in physiology and 
disease came from testing synthetic antimalarials during and espe- 
cially after World War II. Occasional individuals proved "sensi- 
tive" to the antimalarial drugs, that is they developed haemolitic 
anemia due to the destruction of red blood cells. The character- 
istic symptoms were demonstrable in occasional individuals who 
had been given primaquine (a quinine-like drug) such common 
drugs as acetanlid — often used in cold remedies — and the sulfanil- 
amides. A small proportion of individuals maintained on doses 
of primaquine or acetanlid exhibited drug sensitivity but the vast 
majority of individuals did not. 

When the proportion of drug-sensitive individuals was investi- 
gated in different geographical races, interesting results immedi- 
ately appeared. The incidence of multiple-drug sensitivity in 
American "whites" proved to be low, 1% or even less. Among 
1000 prisoners treated with primaquine and described by Dern, 
Beutler and Aving ('55) the incidence of sensitivity was 0.1%. 
However, a larger proportion of American Colored prisoners de- 
veloped haemolitic anemia following the drug treatment. Among 
them, proportions of sensitive individuals ranged from 5% in one 
test to 11% in another group and 12% in yet another group. The 
true incidence of drug-sensitivity in the American Colored popu- 
lation is therefore close to 10% as shown in the following table. 


Per Cent of 
Group Drug Sensitivity 

491 "white" patients Sulfanilamide 1-3%* 

131 Negro patients Sulfanilamide 12.0% 

1000 "white" convicts Primaquine 0.1% 

199 Negro convicts Primaquine 11.0% 

1 Calculated by Beutler et al ('57) based on data by Wood. 


Studies on the blood of drug-sensitive and drug-insensitive in- 
dividuals have shown marked differences in the stability of blood 
cell glutathione. To quote Beutler, Robson and Buttenwieser 
('57) "When primaquine was administered to non-sensitive sub- 
jects there was no change in the red cell GSH (reduced gluta- 
thione ) level. When primaquine was administered to a sensitive 
subject, however, there was an abrupt fall in the GSH content of 
the red blood cell to about one-half of the original already sub- 
normal value." Thus the GSH content of sensitive red blood cells 
is less to begin with, and falls further upon administration of 
primaquine or some similar drug. 

While the haemolitic anemia following drug administration to a 
sensitive subject is ordinarily of minor concern to the general 
practitioner, the far greater incidence of drug sensitivity in Ne- 
groes must be considered. The protection of Negro troops in 
malarial areas becomes a complicated matter, since approximately 
10% of them are sensitive to the drug that protects them against 

Moreover, the possible relationship of this primaquine type of 
drug sensitivity to endemic malaria becomes important to con- 
sider. If the drug-sensitive individual already has a reduced 
oxygen-carrying capacity in the blood, high parasitization for 
Falciparium malaria may not take place. As with the abnormal 
hemoglobins, his erythrocytes simply may not maintain the micro- 
organisms at a critical phase in their reproduction. Thus the drug- 
sensitive person may be relatively immune to malaria, and the 
differential distribution of this trait in individuals of African an- 
cestry may be a clear indication of both where and how the trait 
arose, as Motulsky (1960) has shown. 

Primaquine-sensitivity is caused by a deficiency of the red-cell 
enzyme glucose-6-phosphate dehydrogenase. This type of defi- 
ciency has been shown to be particularly common where malaria 
exists. From Spain to the Philippines the incidence of this hered- 
itary sex-linked blood-enzyme deficiency matches the prevalence 
of malaria. As with the abnormal hemoglobins, primaquine- 
sensitivity marks the impact of malaria on racial genetics. 



Another disorder involving red-cell destruction is Favism. 
Favism is an allergic-like response to the broad or Fava bean. 
Susceptible individuals may develop hemolytic anemia not only 
by eating a dishful of broad beans, but even by walking through 
a field when the plants are in flower. Thus, the responsible agent 
for Favism is present in the pollen as well as in the bean itself 
as Rosen and Scanlan (1948) have observed. 

The disease itself has been recognized for centuries. Its known 
distribution is limited to peoples of Mediterranean origin, to 
Spaniards, Italians, Greeks, Armenians and Jews. But the inci- 
dence of Favism is unknown because of varying susceptibility. 
An individual, not previously affected, may develop symptoms 
unexpectedly after eating the broad beans. Recent studies in 
Israel show that Favism is inherited as a sex-linked trait. 

Similarity between Favism and the drug-induced hemolytic 
anemias suggested a common biochemical mechanism. In Italy 
Sansone and Segni (1957) have shown that blood glutathione 
levels in subjects predisposed to Favism diminished during the 
episodes characteristic of the disease. In Israel, Szeinberg et al. 
( 1958 ) similarly confirmed the evidence that Favism is associ- 
ated with glucose-6-phosphate dehydrogenase deficiency. In the 
Fava-prone individual blood enzyme levels may be materially 
decreased during an attack. 

Under investigation is the relationship between Favism and 
primaquine-sensitivity, the question being whether the same gene 
is involved or whether this represents yet another biochemical 
polymorphism. Of great interest is the fact, observed by Motul- 
sky ( 1960 ) that Favism also appears to be associated with malar- 
ial areas. It is, interestingly, characteristic of Sephardic and not 
Ashkenazic Jews. 

Clearly, malaria has left its impact on many populations 
through numerous adaptive polymorphisms. Favism, in particular, 
provides the opportunity for further studies: one would like to 
know why the incidence of a primaquine-type of sensitivity is so 
high in Kurdistani Jews. Assuming that the gene is inadaptive in 
a non-malarious area, one should be able to compute the rate of 
genetic change for numerous Mediterranean populations (Arme- 


nians, Syrians, Greeks, Italians and Jews) that have migrated to 
northern Europe, North America and the non-malarious regions 
of South America. 


As the reader of this book undoubtedly knows, a great many 
diseases have been considered to be either race-limited or popula- 
tion-limited at various times. For a few, the evidence stands. 
Tay-Sachs disease and familial dysautonomia are both more com- 
mon in Jews. Leprechaunism seems to be restricted to people of 
Irish descent. Silferskiold's disease apparently is a Scandinavian 
specialty. Leukemia appears to be inherently more common in 
Japanese, a matter of considerable importance in ascertaining the 
effects of radiation in Hiroshima and Nagasaki. 

For some of the data apparently linking race and disease, social 
and economic explanations suffice. Alcoholism and Irish ancestry 
are not likely to be related on a purely genetic basis. Rheumatic 
heart disease has been common among each immigrant group in 
the United States in succession. Insanity, despite possible genetic 
predisposition often indicates the working of learned behavior, 
with immigrants adopting the typical psychotic directions of their 
hosts within a generation. 

With other disorders such as coronary artery disease, the pic- 
ture becomes more complicated. We know that the Bantu and 
Zulu, previously coronary-free, do develop coronary artery disease 
as they become Europeanized. Yemenite Jews in Israel were once 
coronary-free : now they resemble European Jews in their growing 
predisposition to coronary disease. 

Diabetes too, once a "European disease" and especially common 
among European Jews, now turns up frequently among the de- 
tribalized natives of South Africa, among Japanese professionals 
and among East Indian intellectuals. To list such diseases as 
exclusively "hereditary" therefore does violence to observed fact, 
and ignores the role of nutrition. 



Nevertheless, such a disease as coronary heart disease does 
have a demonstrable genetic component, in the simply-inherited 
metabolic disorder hypercholesterolemia. Hyperurecemia, simi- 
larly inheritable, is one factor behind the painful disease of gout. 
Some proportion of diabetics are genetic diabetics, in whom the 
disorder may not appear in the absence of particular nutritional 
or emotional stresses. 

The continuance of these obviously disadvantageous disease 
genes, indicates either a high mutation rate, or better, some equiv- 
alent advantage associated with the genes. Since infectious dis- 
eases have long been with us, it is reasonable to look to them 
for the explanation. 

Besides malaria, one may suggest tuberculosis, yaws and 
syphilis, leprosy and elephantiasis, and bihlzaria. In addition, 
there are the childhood diseases — measles, mumps and whooping 
cough, scarlet fever, diphtheria and poliomyelitis. There are the 
ECHO and Coxsackle viruses, debilitating in the adult but often 
fatal in the newborn. 

Within the past decade we have seen the rise of DDT-resistant 
flies and penicillin-resistant staphylococcus. Such genetic immu- 
nity need not be restricted to insects and bacteria. In fact, we 
have evidence that genetic immunity to toxins and bacteria does 
arise in man, and that some of the "hereditary diseases" represent 
adaptations to present or past epidemics. 


The investigation of hereditary diseases in different race- 
populations increasingly supports the contention that the genetic 
characteristics of every race are shaped by the environment in 
which it lives. 

While purely chance events may be responsible for differing 
frequencies of very rare disease genes, gene frequencies of 0.1 and 
higher strongly suggest the workings of natural selection, and 
selection coefficients far higher than those considered plausible a 
fews years ago. 

Because of environmental differences, a particular heterozygote 
may be advantageous in one area and disadvantageous in another 


area, thus leading to differences between local races. However, 
local races from different geographical races may resemble each 
other in respect to particular gene frequencies, if subject to the 
same direction of selection. 

Extrapolating from such diseases as sickle cell anemia, Medi- 
terranean anemia, Favism and primaquine-drug-sensitivity, all 
race-limited hereditary disorders must now be considered as 
possible examples of adaptive polymorphism in man. 


Bennett, G. H., Rhodes, F. A. and Robson, H. N.: A possible genetic 
basis for Kuru, Am. J. Human Genet., 11 .169-187, 1959. 

Beutler, E., Robson, M. and Buttenwiesen, E.: The glutathione in- 
stability of drug-sensitive red cells, J. Lab. and Clin. Med., 49:84-59, 

Dern, R. J., Beutler, E. and Alving, A.: The hemolytic effect of prima- 
quine, /. Lab. and Clin. Med., 45:30-39, 1955. 

Gajdusek, D. C. and Zigas, V.: Kuru, Am. J. Med., 26.442-469, 1959. 

Heller, H., Sohar, E. and Sherf, L.: Familial Mediterranean fever, Arch. 
Intern. Med., 102:50-71, 1958. 

Motulsky, A. G.: Metabolic polymorphisms and the role of infectious 
diseases in human evolution, Human Biol, 32:28-62, 1960. 

Neel, J. V.: The study of natural selection in primitive and civilized 
human populations. Human Biol., 30:43-72, 1958. 

Riley, C. M.: Familial autonomic dysfunction, /. Am. Med. Assn., 
149:1532-1535, 1952. 

Rosen, A. P. and Scanlan, J. J.: Favism, N.E.J. Med., 239:367-368, 

Sansone, G. and Segni, G.: Sensitivity to broad beans, The Lancet, 
273:295, 1957. 

Szeinberg, A., Asher, Y. and Sheba, C.: Studies on glutathione stabil- 
ity in erythrocytes of cases with past history of Favism or sulfa- 
drug-induced hemolysis, Blood, 13:348-358, 1958. 



Jlollowing inexorable laws of genetics, gene fre- 
quencies may be expected to remain constant from generation to 
generation. If the gene frequency for blood group O is 0.64 in 
this generation, it will be 0.64 in the next. If 18% of Englishmen 
have light hair today, 18% had light hair during Victoria's reign. 
Stability of gene frequencies, once panmixia is achieved makes 
possible mathematical excursions into the past, and prediction of 
future trends. 

But gene frequenices do not always remain constant, thus 
changing the genetic nature of a race-population. One mecha- 
nism capable of changing gene frequencies is, of course, natural 
selection, which alters the balance of alleles in one direction or 
another. Mutation is a second mechanism that affects gene fre- 
quencies — introducing new genes into the population through 
mutating (i.e., changing) some proportion of the old. And a 
third evolutionary mechanism, called random genetic drift or 
more simply "drift" is also capable of altering gene frequencies 
but in a random or nondirected fashion. 

Drift, in fact, is largely statistical variation — purely random 
fluctuation of gene frequencies from generation to generation. As 
any student of statistics knows, frequencies or proportions will 
differ somewhat in successive sub-samples as a result of sampling 
error, that is accidents of sampling. In successive human genera- 
tions chance events may slightly lower a gene frequency, or raise 
it unpredictably and in no regular pattern ( see Figure 18 ) . 

Characteristically, genetic drift is of minimal importance in 
large populations. Where the number of individuals, or better 
the number of individuals of breeding age is large, say 1000 or 
more, losses of genes in some lineages are balanced by gains in 






Fig. 18. The mechanism of genetic drift. In small populations or with low 

gene frequencies regardless of population size, the frequency of particular 

genes may vary considerably from generation to generation. Some of the 

differences between local races may be attributed to this process. 


others. In such a large population, random variations in gene fre- 
quency ordinarily will not exceed ± 1%. In small populations, 
however, random variations in gene frequency may loom large. 
With a breeding population of but 100, a 5% increase or decrease 
in gene frequency may easily occur from generation to generation. 
In extreme cases, as where a particular gene is rare to begin with 
or where chance variations run in the same direction for several 
generations, a gene may be eliminated from the population on a 
purely chance basis. 

Moreover, such small populations as have been described are 
by no means small as human groups go. An Australian horde may 
number 400, but excluding grandparents and children, the size of 
the breeding population is less than 100. For a Bushman band, 
and for some Eskimo isolates the breeding population may be 
ten or even less. One can easily imagine the fate of comparatively 
rare genes in such a population — either becoming eliminated en- 
tirely, or the frequency becoming "fixed" at 1.00, due entirely to 
chance events. 

Thus it is that some students of human population genetics, 
among them Gabriel Lasker and Joseph Birdsell, have been inter- 
ested in random drift as explaining some differences between 
some isolates, differences between micro-races, and even differ- 
ences between geographical races. If a gene frequency became 
fixed at either 1.00 or 0.00 just at the time a particular isolate 
began explosive population growth, such a gene frequency might 
characterize the descendants, ultimately numbering in the thou- 
sands. And, for rare and uncommon genes barely represented 
in a population of any size, random genetic drift may well account 
for the population differences that we know. 

There is much attractiveness in genetic drift as an explanation 
for some part of racial differences. After all, most human popula- 
tions have been small, Pleistocene populations rarely exceeded 50, 
equivalent to a breeding population of little more than ten. Neo- 
lithic communities of a hundred or so were equivalent to breeding 
populations of twenty-five. Only rarely, and not much before the 
present era, were populations larger than 2000 practicable and 
these populations were still "small" in terms of the number of 
breeding pairs. 


As migrants migrated, they broke into small groups. The boat- 
loads of Polynesians viewed by Captain Cook constituted small 
populations, ideal for the mechanism of drift. Eskimo groups of 
not more than 50 wandered from Russian Siberia to American 
Alaska, and back again. Even the recent migrations into America 
have involved Lebanese clans, Irish towns, and Armenian vil- 
lages. Such units, detached from the larger group, have undoubt- 
edly been subject to genetic drift, as Bentley Glass has attempted 
to demonstrate for the "Dunkers," a religious isolate of German 
origin. "Drift" may account for some of the differences between 
Yemenite and Alexandrian Jews, or the Jews of North Africa and 
those of Poland or Lithuania. 

Yet there are arguments against according drift too large a role 
in racial differentiation. Drift, being purely chance, does not dis- 
tinguish between adaptive genes and inadaptive genes. Drift 
alone would pile up high gene frequencies for inadaptive genes, 
but such high frequencies would then be whittled down by natu- 
ral selection. Drift could account for some differences between 
adjacent populations, but not differences that are distributed in a 
regular way forming "clines." Only for perfectly neutral genes 
could drift operate alone to bring about major differences, and 
at the present time we are increasingly skeptical of such neutral 
genes, as is evident throughout this book. 

Nevertheless, drift could operate to lose a gene completely, 
and this may be the explanation for the virtual absence of blood 
group B in the Americas. Alternatively, drift in conjunction with 
natural selection could operate to speed up the direction of evolu- 
tion. Given a half dozen populations all subject to the same selec- 
tive forces, the population in which drift and natural selection 
operated together would have the best chance of survival. In 
such a population the largest number of individuals would come 
to possess the optimum gentoype, the fastest, and thus would 
soon outnumber the others. 

Drift can now be investigated in two ways. It is possible to set 
up a computer program to determine the circumstances under 
which random genetic drift may have the maximum race-making 
potentiality. Alternatively, it is possible to investigate drift in 
small human isolates, over time, or in relation to the populations 


from which they have been drawn. While these mathematical 
and investigative techniques can indicate the theoretical impor- 
tance of drift, the extent to which drift actually has been opera- 
tive in bringing about racial differences is extremely difficult to 
determine. Moreover, a small degree of gene-flow from the out- 
side could well counter the direction of drift, through its effect 
on the size of the total breeding population. 


Random genetic drift, the Sewall Wright effect, most commonly 
called "drift" may be considered as a third evolutionary mecha- 
nism responsible for racial differentiation. Drift operates at maxi- 
mum effectiveness in small populations or at very low gene fre- 
quencies. Since most human populations in the past were small, 
ideal in size for drift to operate, this mechanism may explain 
differences between micro-races, local races and even geograph- 
ical races. Nevertheless drift could not operate for long in opposi- 
tion to natural selection. Most likely, drift has been effective 
where it has coincided with the direction of selection in particular 
race-populations . 


Birdsell, J. B.: Some implications of the genetical concept of race in 
terms of spatial analysis, Cold Spring Harbor Symposia on Quantita- 
tive Biology, 15:259-314, Cold Spring Harbor, The Biological Labora- 
tory, 19^0. 

Glass, B. Genetic changes in human populations especially those due 
to gene flow and genetic drift, Advances in Genetics, 6:95-139, New 
York, Academic Press, 1954. 

*Glass, B., Sacks, M. S., John, E. F. and Hess, C. Genetic drift in 
religious isolate: an analysis of the causes of variation in blood group 
and other gene frequencies in a small population, Am. Naturalist, 
86:145-159, 1952. 

Lasker, G. W. Mixture and genetic drift in ongoing human evolution, 
Am. Anthropol, 54:433-436,1952. 

* Lasker, G. W. Human evolution in contemporary communities, 
Southwestern Anthropol, 10:353-365, 1954. 


Wright, S. Classification of the factors of evolution, Cold Spring Har- 
bor Symposia on Quantitative Biology, 20:16-24, Cold Spring Harbor, 
The Biological Laboratory, 1955. 

Wright, S. Fisher and Ford on "The Sewall Wright Effect," Am. 
Scientist, 39:452-458, 1951. 

Reprinted in Readings on Race. 



Ahroughout human history countless isolates have lost 
their reproductive barriers and in consequence their genetic 
uniqueness. Remote tribes have met and exchanged members of 
marriageable age. Larger groups have engaged in warfare, claim- 
ing women from the defeated as concubines and slaves. In the 
long course of human migration dozens of aboriginal populations 
have been swept up in the flood-tide of population expansion, 
joining the winners genetically as well as politically. Victorious 
soldiers have not only contributed genes to the conquered nations, 
but through war-brides brought home have altered the genie 
makeup of their own groups as well. 

In the days of foot-warfare and slow ox-teams and carts, ad- 
mixture was largely between adjacent isolates. When the Israel- 
ites overran the Canaanites and the Philistines, they were among 
peoples much like themselves. Warfare among the Greek city- 
states or between the Romans and their Italic neighbors similarly 
involved micro-races or local races at most. But with the advent 
of well-supplied legions, with the introduction of navies, and with 
all means of rapid mass transport, admixture came to involve 
geographical races as well. 

Today, we see numerous examples of crosses between disparate 
geographical races, with an increasingly larger proportion of the 
world's population so formed. Much of Central and South Amer- 
ica is "mixed," Amerindian times European in origin, and often 
partly African. The American Negro is properly European times 
African as are the Cape Colored of South Africa. Hawaii has long 
encompassed European-Asiatic and European-Asiatic-Polynesian 
mixtures, and European- Asiatic crosses are increasingly frequent 
in North America now. 



Race mixture, whether between local races or geographical 
races, must be added to natural selection and drift as a race- 
making mechanism. Although race mixture does not add new 
genes (as mutation does) and does not of itself remove genes 
from the population, it increases genetic variability, it results in 
new genotypic combinations and thus provides new grist for the 
evolutionary mill. 


Because race mixture of any kind involves a union of separate 
gene pools, each differing somewhat from the other, the immedi- 
ate effect of admixture is to increase genetic diversity. This is 
least true, of course, where isolates are very like each other in gene 
frequencies, and most apparent where previously-separate groups 
differ largely in genetic makeup. In either extreme, however, in- 
creased genetic diversity provides more material for natural selec- 
tion to work upon. Race mixture increases population variability 
and potentially at least, speeds up natural selection. 

But race mixture is far more than the arithmetic addition of 
alleles. While making the new gene pool more complex than was 
true of either parental grouping, admixture also results in new 
genotypic combinations. As a simple example, suppose that one 
group contained the genes a and b, while the second group con- 
tained the genes c and d. In the hybrid population we would then 
find a total of ten different genotypic combinations, aa, ab, ac, ad, 
bb, be, bd, cc, cd and dd. Not only is morphological variability 
increased as a result of admixture, but new genotypes are thus 
produced (Fig. 19). 

The existence of these new genotypes is particularly important, 
because natural selection operates upon genotypes rather than on 
individual genes, and the new genotypes may offer advantages not 
present in any of the parental genotypes. By way of example, the 
genotypes be and bd may ensure greater cold resistance or supe- 
rior heat resistance, or they may protect the possessor against a 
greater range of thermal extremes. In fact, the new genotypes be 
and bd may be so superior to the original genotypes aa and bb 
or cc and dd as to result in a marked change in the genetic nature 
of the hybrids. In this event the products of admixture would 



Fig. 19. Race mixture. Although race mixture does not contribute new 

genes it ordinarily results in new genotypic combinations not found in either 

parental groups as diagrammed above. See Figure 23 for an example. 

eventually provide little clue as to the parental genotypes or to 
the proportions of each group that originally entered into admix- 

Were we to enjoy a long-term view of racial evolution in any 
part of the world, we would witness an almost rhythmic succes- 
sion of steps. First there would be the isolates, each polished to 
a peak of adaptive fitness in its own ecological zone. Then there 
would be admixture, followed by increased genetic diversity, new 
genotype combinations and ultimately new peaks of fitness. Again, 
there would be admixture and further selection among the new 
genotypes, leading to new adaptive peaks again and again and 


again. While we, with our birds-eye time-lapse view could see 
the creative power of admixture and the recurrent development 
of new adaptive modes, the populations involved would be con- 
scious only of the short-term run of events. At each stage, and 
with a myopic view of history, they would loudly proclaim their 
own racial "purity." 


In the last century race-mixture, or more specifically mixture 
between geographical races was commonly viewed as harmful. 
Europeans, themselves the product of centuries of admixture be- 
tween local races, took a dim view of the new race-hybrids their 
own expansion produced. Arguing on quasi-biological grounds, 
particularly with reference to the mule, impaired fertility was said 
to be one product of human racial miscegenation. Human "hy- 
brids" of various sorts were claimed to be indolent, immoral, un- 
trustworthy and uneducable. Progeny resulting from race mixture 
were alleged to fall below the intellectual level of either parent 
race. Somewhat inconsistent however, was the attitude toward 
Eurasians. As pictured in mystery novels as late as 1920, Eur- 
asians were exceptionally sinister, sly, Machiavellian, yet capable. 
When a Eurasian entered an Edgar Wallace novel, shutters 
clanked, doors squeaked, and the heroine was tethered in a junk 
bound for Singapore with the pure-English hero in close naval 

Despite such novel views, human "hybrids" ( that is geographi- 
cal race-crosses ) have evidenced no signs of impairment. Popula- 
tion expansion in Middle and South America belies any diminu- 
tion in fertility. No evidence exists as to reduction in mental 
acuity. Morality among products of race mixture has been neither 
lower nor higher than their station in life allows. In Hawaii 
(as elsewhere) Eurasians have not lived up to their reputation 
as sinister, but have contributed lively, useful, imaginative and 
law-abiding citizens, many of them unusually attractive as the 
motion-picture audience has come to discover. 



While exponents of racial purity were still arguing the bio- 
logical inferiority of human hybrids, and were suggesting reduced 
fertility, impaired viability and psychological instability as the 
price of miscegenation, plant experimenters encountered a totally 
opposite trend. Intentional plant hybrids, made by crossing dis- 
tinct strains or "races," proved to be superior in fertility, superior 
in growth and superior in disease-resistance. This phenomenon, 
still imperfectly understood, is the result of crossing genetically- 
distinct lines. Limited to hybrids, especially Fi (first generation) 
hybrids, it has been termed hybrid vigor. 

One conspicuous example of hybrid vigor is hybrid corn. In the 
field the stalks are taller and sturdier and the ears larger, more 
numerous and better filled. Hybrid tomatoes are another ex- 
ample: the premium "hybrid" seeds cost more, they produce 
bigger and more vigorous plants and finer, firmer, better-fleshed 
tomatoes. Intentional hybrids in other lines evidence greater dis- 
ease resistance, and if well-fertilized, amazing productivity. One 
explanation for this phenomenon of hybrid vigor is the dispersal 
of deleterious genes in the Fi generation. Notably, the advantages 
are less obvious in the F 2 back-crosses, so that fresh hybrid seed 
must be used at each sowing or planting. 

Does hybrid vigor exist in man, in the first filial generation from 
geographical race-crossings? Here we encounter an experiental 
problem, that of controlled conditions. Whereas hybrid and 
straight-line corn can be tested in the same field, under compa- 
rable conditions of temperature, rainfall and fertilizer, such ex- 
perimental controls have not been possible for man. Where can 
we compare, except possibly in Hawaii, Asiatic European and 
Asiatic-European crosses under comparable conditions of nurture? 

Some few human populations have been offered as examples of 
hybrid vigor. The tall, vigorous, but carious Pitcairners, descend- 
ants of the Bounty mutineers and their Tahitian wives, have been 
offered as an example of hybrid vigor in man. But these living 
Pitcairners are not FiS, but complex back-crosses, primarily of one 
male line. In like fashion, the Norfolk islanders are not first gen- 
eration hybrids. Something other than hybrid vigor, possibly 
natural selection, possibly adherence to a rather English way of 


life, accounts for the size and ruggedness of these descendants of 
the Bounty studied thirty years ago by the anthropologist H. L. 

Still, if "hybrid vigor" exists in man, we, should be able to detect 
it under circumstances where environmental variables are well 
controlled. By way of example, growth studies of English and 
English- Negro infants in Liverpool orphanages may well provide 
information on hybrid vigor. Alternatively, instead of considering 
first generation geographical race-crosses we may investigate 
highly inbred human lines, where suppressor genes may limit size 
and growth. Then, by measuring the progeny of out-marrying 
members of such isolates, size increment due to "heterosis" may 
well be demonstrated. 

Such an investigation has been conducted by Dr. Frederick 
S. Hulse. For his "inbred" population, he used members of iso- 
lated Swiss cantons where cousin marriage was common. His 
"outbred" comparison group involved members of the same can- 
tons who married outside of the canton. And, as shown below, 
the progeny of the out-marrying Swiss proved larger than did 
the children of the still-inbreeding Swiss. The experimental 
design, using intra-canton and inter-canton matings, shows how 
environmental circumstances can be kept relatively constant, 
thus avoiding the major pitfall of most attempts to uncover 
hybrid vigor in man. 







Stature (cm) 



Weight (kg)) 



Shoulder breadth (cm) 



Head length (mm) 



* From Hulse, F. S.: Exogamie et heterosis. Arch. Swisses d' Anthropologic generate, 
22: 103-125, 1957. 



Though there are obvious difficulties in the study of race mix- 
ture, chiefly arising from the lack of controlled conditions, there 
is much to be gained from such investigations, particularly in the 
area of human genetics. The mode of inheritance of many human 
traits can best be investigated in hybrid progency. Studies of 
Fi and F 2 hybrids can help to indicate the degree of complexity 
of polygenic traits such as skin color, hair form, body proportions 
and the like. Such a characteristic as the tight spiral-tufts of 
Bushmen and Hottentots can not be analyzed genetically in 
Bushmen or Hottentot, all of whom have it, or in Europeans who 
lack it. But in various Hottentot-Boer "crosses," the genetics of 
spiral-tuft hair form can best be tackled. 

Thirty years ago Caroline Bond Day collected photographs, 
hair samples and pedigrees of Negro-white families. Her work 
helped to indicate the genetic complexity of hair form, and to 
show that skin pigmentation was controlled by multiple genes, 
involving at least three loci. Studies on (aboriginal) Australian- 
European crosses and Asiatic-Australian crosses have further 
illuminated the phenomenon of graded dominance, with a par- 
ticular gene acting as a dominant in one type of crossing, and as 
a recessive in another type of crossing. Human hybrids serve a 
useful purpose in the analysis of many genetic traits, and make up 
in part for our inability to institute experimental matings of the 
kind useful in plant and animal genetics. 

Hybrid populations also afford the possibility of searching 
for linkage, something far less practical in old-established groups 
where linkage has been disrupted by crossing-over of chromo- 
somal parts. The geneticist David Rife has conducted studies of 
this kind, in East- African populations of known and recent hy- 
brid origin. 

Finally, hybrid populations afford the possibility of calculating 
the degree of admixture, a useful accomplishment since historical 
sources are rarely quantitative. Available evidence, as investi- 
gated by D. F. Roberts and others, gives approximately 20% of 
"European" genes to the American Colored population. However, 
and contrary to popular opinion, the American Colored population 
contains few Amerindian genes, as Bentley Glass has demon- 



strated. Thus, the myth of considerable Amerindian admixture 
is exploded. Clearly such studies as those by Roberts or Glass 
prove the value of investigating geographical race hybrids (see 
also Fig. 20). 


In the last century race-mixture was viewed as totally bad, with 
dire consequences visited upon the hybrid progeny. Plant genet- 
icists, however, discovered the phenomenon of hybrid vigor, thus 
pointing to a possible advantage of hybridization though such 
has by no means been conclusively demonstrated in man. What 
can we say now about the advantages and disadvantages of 
hybridization, or race-crossing in Homo sapiens? 









Serological and morpholog- 

Fig. 20. Mathematical analysis of race mixture, 
ical data give slightly different results when the number of variables con- 
trasted is small. With a large number of variables the two methods would 
be expected to yield identical results. (From Pollitzer, 1958.) 


Taking the long view and centering our attention on the popu- 
lation, hybridization may be considered as advantageous. By 
increasing genetic variability, there is a greater range of geno- 
types to work from, a greater likelihood of adaptive genotypes 
and far better prospects for long-term survival. In the long run, 
the more different genes the better, and race-crossing enhances 
genetic diversity. A hybrid race has superior long-range prospects. 

The short-range view, however, centering attention upon in- 
dividuals rather than the population can be somewhat different. 
If a population is optimally adapted to the environment, the 
introduction of new genes will lower average individual fitness, 
until the balance is restored by natural selection. But this answer 
itself needs qualification and emendation, as shown in the follow- 
ing example. 

Take an East African population constantly beset by malaria 
and with a high incidence of the sickling gene. This population 
is optimally adapted to its circumstances. Admixture with non- 
sickling peoples would result in decreased average fitness, until 
such time as the balance is restored by selection. For this par- 
ticular example, admixture is bad for the individuals concerned. 

If we move this African population (or control malaria with 
DDT, ditch drainage and quinine) the situation may become 
quite different. The "hybrids" with a lower incidence of the 
sickling disease would exhibit increased fitness, relative to the 
original population. 

So the "fitness" of hybrids depends very much upon circum- 
stances. Given a population highly adapted to particular cir- 
cumstances, and an intrusive population not, the hybrids would 
be less fit than one parental group (though more than the 
second). Change the circumstances and the hybrids may be 
superior in fitness to the first group but not the second. In a 
third set of circumstances hybrid fitness may be superior to both 
parental groups. 


Race mixture, the coalescence of distinct micro-races, local 
races or geographical races is an old human accomplishment re- 
sulting in increased genetic diversity, more rapid natural selection 
and (usually) new adaptive peaks. 


Apart from the advantages accruing to scientific investigation, 
race-mixture raises the problem of population and individual 
fitness. No hard and fast rule can be drawn. In some cases 
decreased individual fitness ensues, and in other cases increased 
individual fitness will eventuate. From a population point of 
view, increased genetic diversity is of ultimate value. 

Dire predictions about race mixture voiced in last-century 
Europe have not proved correct. Decreased fertility in par- 
ticular has failed to evidence itself in hybrid human populations. 
Hybrid vigor, suggested by plant experiments, remains to be 
demonstrated in man. 

That race-mixture has been an important race-making mecha- 
nism goes without doubt. Studies in populations of recent, hybrid 
origin hold many advantages and there are many opportunities 
to investigate fitness under purely local circumstances. 


Day, C. B.: A Study of Some Negro-White Families in the United 
States, Cambridge, Peabody Museum, 1932. 

*Glass, B.: On the unlikelihood of significant admixture of genes from 
the North American Indians in the present composition of the Ne- 
groes of the United States, Am. J. Human Genet., 7:368-385, 1955. 

Pollitzer, W. S.: The Negroes of Charleston (S. C), a study of hemo- 
globin types, serology and morphology, Am. J. Phys. Anthropol, N.S. 
i6:241-263, 1958. 

* Roberts, D. F.: The dynamics of racial intermixture in the American 

Negro — some anthropological considerations, Am. J. Human Genet., 

Shapiro, H. L.: Descendants of the Mutineers of the Bounty, Honolulu, 
Memoirs of the Bernice P. Bishop Museum, vol. 9, 1929. 

Shapiro, H. L.: "Race Mixture," in Anon.: The Race Question in Mod- 
ern Science, New York, Morrow, 1956. 

Stuckert, R. P.: African ancestry of the white American population, 
Ohio J. Set, 58:155-160, 1958 

* Reprinted in Readings on Race. 



jl\. century ago most thoughtful and learned individ- 
uals firmly believed in racial differences in temperament. The 
stereotypes of the past century, sometimes still reiterated as fact, 
painted vastly different natures for the various geographical races. 
Asiatics were pictured as sly, mysterious and inscrutable and 
Africans were described as childlike and improvident. American 
Indians, with some exceptions, were portrayed as stolid, stoical 
and humorless. The relationship between geographical race and 
temperament, from a 19th century European view, may be sum- 
marized by one quotation: "Black men are ruled by passion, 
yellow men are bound by custom, while white men are governed 
by law." 

Even within geographical races, inborn differences in tempera- 
ment were accepted as fact. It is part of our literary heritage to 
speak of melancholy Danes, musical Italians and volatile French- 
men. Various nations have been accused (by various other na- 
tions) as being humorless. Undesirable characteristics have been 
attributed by one group to the other. The Englishman speaks of 
"taking French leave," while the French equivalent is to "take 
English leave." 

To some extent, these stereotypes were based on inadequate 
observations, and were due to unfamiliarity with language and 
gesture. It is difficult to interpret the facial expressions of na- 
tionals from another country and still more difficult to operate in 
a totally different culture. At the same time some stereotypes are 
factual. Italians do gesticulate. Germans do eat wurst. And the 
Sioux Indians were notably inhospitable to General George 
Custer. Such stereotypes, however, neglect to distinguish learned 



behavior from inherent nature; they do not distinguish being 
Chinese from speaking Mandarin, nor a hundred generations of 
trading experience from inherited mercantile ability. 

It is, moreover, vastly interesting to see how stereotypes change 
with time, The noble and restrained Romans became the voluble 
and musical Italians without major genetic change. The patri- 
archal Hebrews became the peddlers and merchants despite little 
accretion of outside genes. The woad-painted, skin-draped bar- 
barians near Londunum, of Augustus' time, became the chief 
carriers of western culture two millennia later. The same English 
yeoman stock gave rise to the nasal Yankees, the elegant Vir- 
ginians, the backwoods Appalachians and the Texans. 

Clearly, the outstanding character-traits that do distinguish 
groups are culturally-determined and culturally formed. Associa- 
tions between occupation and nationality melt like wax with 
changes in social mobility, with improved education, and with 
altered economic status. Predilections toward criminality, initially 
characteristic of each immigrant group into the United States, 
disappear with succeeding generations. Even the pattern of crim- 
inal proclivity, characteristically different in various national 
groups when they first arrive, rapidly readjusts to the general 
norm as is clearly evident in the table on this page. Given such 
data, claims for true racial differences in criminal preferences 
rapidly dissipate. 



Irish Immigrants 










* From Sutherland, E. H. and D. R. Creasey: Principles of Criminology. Phila- 
delphia, J. B. Lippincott, 1955. 

So far, there is no evidence for racial differences in character 
and temperament, other than those due to cultural conditioning. 
The same may be said for behavior in general. With respect to 
psychotic behavior, however, there may well be some differences. 
Some of the differences noted between Italian and Irish psy- 


chiatric patients may have a genie background. Differences in 
adrenochrome production, in serotonin levels in the brain, even 
in the central nervous system may explain apparent racial differ- 
ences in the predisposition to particular psychiatric disorders. 
Such differences, however, have not been demonstrated on a 
racial basis, holding the way of life constant. 

Still, the most profitable area for investigation would seem to 
lie in the autonomic nervous svstem. There are clear-cut, repro- 
ducible individual differences in autonomic response specificity. 
Individuals exhibit marked and consistent responses to various 
kinds of stress. Racial differences in gene-determined response 
patterns are very likely, especially when one considers the adap- 
tive nature some of these response patterns must hold in nature. 
Racial differences in behavior therefore, may fall to the psycho- 
physiologist to discover, and will not encompass most areas of 
social behavior. 


Much as racial differences in temperament were firmly credited 
in the century now past, marked hereditary and racial differences 
in intellectual capacity were also accepted as self-evident truths. 
Long before Binet's measurements of intellectual accomplish- 
ment were published in 1905, literate and technologically-ad- 
vanced Europeans held low opinions of the intelligence of illiter- 
ate and technologically-simple "natives." What the natives 
thought of the Europeans, unfortunately is not on record. 

Intelligence tests, originally developed and used as guides for 
the grade-placement of school children at first provided powerful 
support for the assumption of racial differences in intellectual 
capacity. Native-born Americans ranked highest. Immigrant chil- 
dren from northwestern Europe came next. At the bottom of the 
list were American Negro children. Since Binet's test demanded 
reading skills and many of the Negroes were largely illiterate, a 
substantial proportion were rated in the dull-normal, dull, border- 
line and subnormal categories. 

With provincial opinions as to the inferior mental capacities of 
"Injuns," "Chinks" and "Niggers" thus apparently confirmed by 
the new science of psychometries, the famous Army Alpha tests 
of 1917-1918 came as quite a shock. In this battery of tests, Ne- 


groes ( as a group ) fared worse than whites ( as a group ) as had 
been expected. But Negro recruits from some northern states 
averaged higher on the Alphas than did white recruits from some 
southern states. Unless the brighter Negroes had streamed north, 
leaving the dullards behind, and unless there had been a com- 
parable southward migration of the least-capable whites, differ- 
ences in intelligence could not be completely racial. While a 
greater degree of admixture might, in theory, explain the superior 
showing of northern Negroes, what genetic mechanism could 
explain the lower average scores for southern rural white recruits? 

Today, it is quite clear that intelligence tests, such as the Stan- 
ford revision of Binet's, or the Wechsler-Bellevue, or the Otis, 
do not primarily measure inherent gene-determined potentialities. 
Intelligence tests are still validated against school performance; 
they measure a "book" kind of proficiency. Such tests do not 
purport to measure many of the skills and abilities that the term 
intelligence commonly suggests. Intelligence tests, moreover, are 
never culture-free. They measure in relation to a particular cul- 
tural background, placing premiums on language skills, urban 
living and acquired knowledge, thus down-rating rural, non- 
verbal or immigrant children or adults. I.Q. test scores neatly 
reflect the level of motivation and (except at the extremes) not 
inherent ability. 

Quite obviously it is impractical to get meaningful comparative 
intelligence scores for many preliterate peoples. Even the culture- 
free test is a misnomer, being a culture-constant test, or else an 
impossibility. Comparing or equating test scores for a Chinese 
peasant boy from Langsai, and a merchant's son from London 
introduces extraordinary problems. Only where education, op- 
portunity and the way of life are uniform across racial barriers 
can meaningful comparisons be made. Possibly comparative 
.testing can be accomplished in Hawaii. Possibly multi-racial 
orphanages may yield the critical data, though institutional con- 
ditions rarely afford the child an adequate opportunity to develop 
the skills most highly rated on the tests. 

Racial differences in measured intelligence thus remain neither 
proven nor disproved. There are differences, but like stature, they 
do not necessarily indicate the maximum level of capacity in the 
absence of standard or controlled conditions. To the confirmed 


believer in racial differences in intelligence, we can simply say 
that the more nearly two groups are matched in educational level, 
family background, opportunity and security, the closer they 
agree on averaged I.Q. scores. To the dedicated equalitarian, the 
believer in no race differences, the disparate levels in the cur- 
rently best-matched Negro-white comparisons stand to be refuted. 

As a matter of opinion, backed by some personal experience 
with mental test data, one may question whether major racial dif- 
ferences in working intelligence could have arisen — except in re- 
mote areas — during the millennia of human evolution. For most of 
man's million-year existence, the way of life of one group strongly 
resembled the life-way of another. Life in Paleolithic England 
and stone-age Tanganyika were much alike. With a salubrious 
climate there arose population pressure and man organized to out- 
smart man. In less equable climes man organized to counter 
nature. One may question whether it took more brains to suc- 
ceed in Neolithic Ireland or in the Indus Valley in the New Stone 

Europe, moreover, was not always the center of technological 
advancement, having attained that status rather late in human 
affairs (and even now the baton is passing). Europeans in fact, 
did not compete with each other on a purely intellectual basis 
until rather recently, but still only a fraction of a percent later 
than Egyptians or Mesopotamians. Besides, one may well wonder 
whether it takes more "intelligence" to survive in Kansas City 
than in the Kalahari. Only if there was long-continued differential 
selection for intelligence in some areas, and not in others, could 
we expect true racial differences in intelligence to exist. 

Yet, it would be a mistake to ignore intelligence completely in 
considering either race or race differences. Different skills have 
had adaptive value in different areas, and in some cases over a 
respectable number of generations. Eskimo mechanical genius 
stands out as one example, and mechanical skill appears to be one 
of the "special" abilities. The exquisite form-color sense possessed 
by many peoples from Japan to Thailand to Burma may well 
represent a second example of special skills, differentially dis- 
tributed with respect to race. 

A very reasonable guess is that races are comparable in the 
sum and total of what we call "intelligence," but differ in many 



interesting details. As with the autonomic response patterns that 
so neatly differentiate one individual from the other, race differ- 
ences may exist in form-discrimination, color-sense, tonal-memory, 
mechanical reasoning, abstract reasoning and with other special 
(rather than general) aspects of intelligence. This supposition, 
moreover, is directly susceptible to testing. 


Racial differences in gesture, speech, emotionality and way of 
life clearly exist but these have proven to be largely differences 
acquired in the social matrix. With migration, acculturation or 
even the simple passage of time, racial "characteristics" change 
immensely as in the case of the rude and crude woad-painted 
natives of colonial Britain. While no racial differences in tempera- 
ment, behavior or intellectual capacity have been firmly estab- 
lished as gene-determined, and while such complexes as overall 
behavior or overall intelligence may not differ from race to race, 
there is every likelihood that components of behavior may be 
gene-determined, with differences in frequency from population 
to population. Here attention may be directed to the patterns of 
autonomic response to stress, to differential skills and abilities 
and to susceptibility to emotional disorders. 


Anonymous: The Race Question in Modern Science, UNESCO publi- 
cations, New York, Morrow, 1959. 

Anastasi, A.: Differential Psychology, New York, Macmillan, 1958. 

Eels, K., Davis, A., Havighurst, R. J., Herrick, V. E. and Tyler, R. W.: 
Intelligence and Cultural Differences, Chicago, University of Chi- 
cago Press, 1951. 

Klineberg, O.: Race Differences, New York, Harper and Bros., 1935. 

Kagan, J. and Moss, H.: Parental correlates of child's I.Q. and height, 
Child Develop., 30:325-332, 1959. 

Lacey, J. I. and Lacey, B. C: Verification and extension of the principle 
of autonomic response stereotypy, Am. J. Psychol., 71:50-73, 1958. 

Moss, H. and Kagan, J.: Maternal influences on early I.Q. scores, Psych. 
Reports, 4:655-661, 1958. 



jfx taxonomy is a listing, an accounting and an enumer- 
ation of the larger and smaller taxonomic groupings for any life 
form. Ultimately, it is concerned with species within a genus, 
and in a specialized taxonomy, with races within the species. 

For man, that is for living man, there is but one species which 
Linnaeus hopefully termed sapiens. Within this species there is a 
relatively small number of geographical races and these can all 
be listed. There is no need to set off certain geographical races 
as the possible result of ancient admixture. The geographical 
races of mankind can be described in relatively short order, there 
being no more than ten of them now, however many more there 
were three thousand years ago. 

Local races, however, pose an obvious problem. There are far 
too many even to list in the very few pages that are available. In 
addition, there is the obvious fact that as many local races have 
yet to be completely identified as we now know. Obviously, some 
selection must be made, if only to acquaint the student with the 
kinds of populations that we define and study. But he also needs 
to know the basis of the selection, why particular race-populations 
that he may never have heard of are included, and why others 
(more familiar) are not even mentioned. 


As repeatedly stated in this book, geographical races are geo- 
graphically delimited collections of local races. Within a geo- 
graphical race overall resemblances are greater than they are 
between geographical races. Nevertheless, in a particular respect, 
resemblances may be greater across geographical barriers than 



between them as, for example, in the incidence of non-tasters 
(cf. Fig. 10). 

Before the evolutionary nature of race was clearly compre- 
hended, non-uniformity of local races within a geographical area 
posed numerous problems. One obvious expedient was to select, 
in each geographical race, a particular local race as the "holotype" 
or representative type. Then, the investigators tried to explain 
why the remaining local races in each area failed to come up to 
the specifications of the holotype, an expedient that gave good 
exercise to the imagination but few contributory results. 

However, there was never any good reason to question the 
ancestry of Africans who were less dark, less prognathous or less 
peppercorn in hair shape than the approved Forest Negro holo- 
type, and there was no reason to set up the "extreme" Mongoloids 
of Northern Asia as the holotype for Asia. The important point 
to bear in mind, in connection with the following listing of nine 
geographical races, is the existence of very considerable diversity 
within each. The fact that local races within each geographical 
race differ from each other is the subject for investigation, one 
of the major reasons for the study of race in man. 

1. The Amerindian Geographical Race consists of a large num- 
ber of local populations, ranging from Alaska, Northern Canada 
and Labrador, to the very tip of South America ( Fig. 21 ) . Often 
marginal, hunting and gathering or semi-agricultural, in most 
areas populations remained small, and genetic isolation was 
frequently complete. Under these circumstances the extent of 
local differentiation found among the aboriginal Americans may 
be taken as representative of Europe and Asia at a much earlier 

Serologically, the Americas are characterized by the low in- 
cidence ( or virtual absence ) of B, the generally low incidence of 
A ( and then only Ai ) , by the low incidence of N, and by varying 
frequencies of the Diego-positive gene Di a . Morphologically, in 
hair form, tooth form, eyelid form, etc., there is an obvious over- 
lapping with Asia and to some extent Polynesia. It is not neces- 
sary to postulate neatly-spaced waves of separate peoples from 
different geographical races to account for American Indian poly- 
morphism. At the same time, some part of the wave theory may 



Fig. 21. Polar-projection map of the world showing the limits of the nine 

geographical races described in the text. In each case geographical barriers 

set off the race-collections. 

well be correct. There were people in the Americas by the end of 
the ice age, and they probably contributed to the ancestry of the 
present Amerindians; the amount, however, is markedly open to 

2. The Polynesian Geographical Race occupies a vast territory 
in the Pacific, ranging from New Zealand to Hawaii and Easter 
Island (Pascua). Many now-uninhabited islands still bear relics 
of this hardy collection of seafaring local races. 

Polynesian polymorphism is marked for hair form, facial fea- 
tures and body build, less so for many serological factors. There 


is a range of skin colors from dark to quite light, and a fair range 
of nose form. Individual Polynesians may be reminiscent of 
Malays, Australoids, even Hopi Indians — but such resemblances 
are no proofs of ancestral relationships. 

Serologically, the Polynesians have a high N/M ratio, as is char- 
acteristic of Pacific peoples, little B, and a high frequency of 
Duffy. Their population movement was primarily west to east, as 
Captain Cook was able to confirm. 

Polynesian polymorphism was explained in terms of a simple 
tri-hybrid origin in the days when a three-race cosmogeny was 
popular. However, serological data are incompatible with such a 
simple explanation and the genetically-naive data analysis that 
favored the tri-hybrid theory has since gone out of fashion. 

3. The Micronesian Geographical Race occupies a series of tiny 
islands in the Pacific, ranging from Ulithi, Palau and Tobi (near 
Guam) to the Marshall and Gilbert Islands. Possessing dark skins, 
wavy, helical or even "frizzly" hair, the Micronesians have been 
alleged to be partly "Negroid" in origin. 

However, the serological picture is, except for the presence of 
blood group B, much like the Polynesian. There is a high pheno- 
type frequency of A, exceeding 50%, and that primarily of the Ai 
subtype. N exceeds M except in the Kapinas. Largely Duffy posi- 
tive ( Fy a ) , and entirely Diego negative, the Micronesians occupy 
a unique serological niche. With R practically absent there can 
be no recent contact with Africa. With a moderate frequency of 
B, the Micronesians are set off from Polynesia and Australia 
while the absence of Di a individuals sets them off from Asia as 

The picture, therefore, is of distinct differences from surround- 
ing geographical races and the virtual impossibility that the 
morphology and serology of the short Micronesians could have 
come about by any simple Negro-European- Australian mixture. 

4. The Melanesian-Papuan Geographical Race isolated until 
W T orld War II, exemplifies the problem of using similarities to 
prove origins. Often owning skulls that are primitive by any 
name, hair that curls, twists, frizzes and occasionally noses that 
seem strikingly reminiscent of the Near East, they are "stone age" 




peoples newly but successfully inured to jeeps and calculators 
(Fig. 22). 

The Papuan-Melanesian peoples are separated serologically 
from the Australians to the south having much more B and 
morphologically and serologically from the Polynesians to the 
east. Isolated from the rest of the world, possibly for 20 centuries, 
they, exhibit a number of gene-determined diseases all of their 
own ( chapter VII ) . 

5. The Australian Geographical Race constitutes a series of 
local races clearly allied with the now-extinct Tasmanians. They 
are big-toothed (often exceeding the classic Neanderthaloids ) 
and they have very long, very narrow skulls, broadest at the base. 

Fig. 22. Melanesians of New Guinea showing dark skin and helical to wooly 
hair. (Photograph, courtesy of Associated Press.) 



Despite their generally-dark exposed-skin color, there is a 
moderate amount of light or red-gold hair, and male pattern- 
balding reminiscent of Europeans. The Australians are notable in 
their very high incidence of N, in the M-N series, in the very low 
frequency of B ( which seems to be recent and adventitious ) . All, 
apparently are Duffy positive ( Fy a ) . 

Some Australian aborigines, even when habituated to clothes, 
are able to withstand very low night temperatures while sleeping 
in the nude. This, however, is a central Australian trait, again 
pointing to the importance of local adaptations within each geo- 
graphical area. 

6. The Asiatic Geographical Race occupies continental Asia, 
and extends also to Japan, the Philippines, Sumatra, Borneo and 
Celebes. It encompasses "the little brown men" of Indonesia 
and southeast Asia, the ruggedly-tall (and often big-eared) 
Tibetans, the North Chinese, the Mongolians and many of the 
natives of Formosa (Taiwan). 

Asia is characterized by blood group B, up to 40%, by Diego 
and (insofar as is now known) the excretion of BAIB. It is the 
continent of inner-eye-folds, broad and fat-padded malars, little 
body hair and sparse beards, and coarse, straight head hair and 
little male pattern-balding. Sexual dimorphism is often limited by 
European standards; and protuberant bosoms and projecting fe- 
male posteriors are rare. The birth rate in Asia, however, shows 
how superficial these characteristics are. 

Judging from local differences, Asia was once characterized by 
hundreds of local races. Even today, at least a hundred can 
easily be distinguished. Since emigration has largely come from 
Japan, South China and the Philippines, we are most familiar 
with the reduced leg/trunk ratio of people from these areas. 
Short legs may in fact be a characteristic of the Asiatic Geographi- 
cal Race, but our samplings are weighted and not really represen- 

It should be emphasized most strongly that there is consider- 
able polymorphism in Asia, extending even to gray-green eyes in 
isolated groups. Moreover, so-called Asiatic or "Mongoloid" char- 
acteristics are rather common over four-fifths of the globe. For 
this reason, an inner eyefold, a projecting malar, 100 mu-wide hair 


or a flat nasal root should not be seized upon as indication of 
"Mongoloid" origins. 

7. The Indian Geographical Race extends from the high Hima- 
layas (from the territory of the Improbable Snowman) to the 
torrid Indian Ocean. Broken into a number of local races, with 
different languages and religions and including castes that are 
true local races in themselves, the Indian Geographical Race 
poses a major task for the excellent anthropologists of India. 

To a European, many Indians look "European." Excepting skin 
color, they often look like natives of the southern and eastern 
Mediterranean. Hair and beard distribution, middle-phalangeal 
hair patterns and parafrontal balding complete these resem- 

Blood group B is often quite high in India (>35%). The Rh 
negative gene (r) is low. These two facts together separate India 
from Europe, and ally it more nearly with Asia. Add to this high, 
often exceptionally high proportions of non-tasters, dark, some- 
times very dark skin color, and such traits as the contiguous eye- 
brows and the uniqueness of this group of populations stands out. 

In the Indian Geographical Race ( and this includes the peoples 
of Pakistan, Kashmir, etc. ) the system of castes both complicates 
racial analysis and points to cultural influences on raciation. While 
the castes in part reflect historical-political stratification, there can 
be little doubt that hereditary occupations (with all of their ad- 
vantages and disadvantages) afford possibilities of differential 
selection within larger demographic units. 

8. The European Geographical Race comprises a collection of 
local races and micro-races inhabiting ( in pre-Columbian times ) 
Europe, Western Asia, the Middle East and Africa north of the 

Serologically, the European Geographical Race is unique in 
the relatively high frequency of the Rh-negative gene. As a 
geographical race it is also unique in the extent of male hirsutism, 
and of male pattern-balding. In pigmentation, however, there is 
considerable overlap among geographical races, though the light- 
est-skinned individuals do come from Europe, primarily northern 
Europe. Sexual dimorphism is often more extreme than in other 


geographical races, though this dimorphism is often exaggerated 
by obesity. 

With its wide geographical range, and considerable poly- 
morphism with respect to skin, hair and eye pigmentation, there 
is undoubtedly a series of local and regional adaptations. In all 
probability northern Europeans are relatively cold-adapted. Sev- 
eral adaptations to malaria are noted in the southern and eastern 
ranges of this geographical race. The high proportion of non- 
tasters in the Baltic probably represents an adaptation to avail- 
able bitter-tasting antithyroid-containing foods. 

Unfortunately, human taxonomy began in Europe and among 
some rather atypical local races. As a result, the northwest Euro- 
pean has been used as reference standard both for modern man 
and for now-extinct fossils. Actually, the European Geographical 

Fig. 23. Spiral-tuft form of the body hair in an American Colored individual. 

This is a phenotypic expression of a genotypic combination not found in 

either parental stock. See Chapter IX. 


Race is by no means particularly "advanced," evolutionarily 
speaking, and undoubtedly perpetuates a fair proportion of late 
Paleolithic genes. 

9. The African Geographical Race constitutes a collection of 
local races and micro-races, all indigenous to Africa north of the 
Sahara. At least some local races exhibit heat-adaptations. A 
variety of adaptations to malaria through altered hemoglobins and 
changed blood enzyme levels are known. 

Serologically, Africa is best characterized by the Rh subtype 
which attains 70% in some areas. Diego-positive individuals on 
the other hand are lacking in Africa. The rare U-negative type 
(in the MN-S system) apparently has its homeland in Africa. 
Africa too, is the distributional center of the sickling trait, the 
Hp 2 haptoglobin type and keloid formation. 

Skin pigmentation is variable throughout Africa but in some 
areas the epidermis and its derivatives are literally stuffed with 
melanin, even to the gums, whites of the eyes and enamel of the 
teeth. However, in East Africa and among the aboriginal in- 
habitants of South Africa, skin pigmentation is considerably less 

Africa includes the extremes of stature, ranging from the true 
pygmies to the tall Nuer. Body build is also variable, and while 
the extreme leg/trunk ratio is common, it is by no means char- 
acteristic of all of Africa. 

The pepper-corn form of the head hair, as typically seen in 
Bushmen, is common in much of Africa but to the north head 
hair is both longer and less neatly spiral or even helical. Increased 
numbers of sweat glands, reported in Equatorial Africa by Hier- 
naux may not be characteristic of the African continent as a 

As with the Asiatic Geographical Race there is no one popula- 
tion representative of the entire geographical race. Even more 
important, dark skin color and helical hair elsewhere in the world 
need not have an African origin. 



In providing a listing of local races, it is obvious that a complete 
enumeration could hardly be given here. Even a summary tabu- 
lation of Amerindian local races would usurp the space allotted 
to this chapter. But ignoring local races would be most undesir- 
able. It would leave the field to the geographical race-collections 
and it would give no indication of the populations that students 
of race actually study. 

Accordingly, a brief list of 32 local races and local race-popula- 
tions has been drawn up. It is obviously not a complete listing 
by any account. The basis of selection here is as follows: 

I. Representative large local races, numbering into the tens of 
hundreds of millions and including numerous micro-races often 
distinguishable as local "types." 

II. Representative small and isolated local races corresponding 
almost perfectly to the idealized population-isolate. 

III. Representative marginal long-isolated local races that have 
been unaltered by admixture for millennia. 

IV. Representative hybrid or mixed local races, formed by 
admixture among different geographical races within the last 
century or two. 

I. Representative Large Local Races ( for location, see Fig. 26 ) . 

1. Northwest European. Comprising the peoples of Scandi- 
navia, northern Germany, the Low Countries, to the United 
Kingdom and Ireland. Variable in size and pigmentation, but 
including a fair proportion of blondism, light eyes and skin and 
a high incidence of O. Including so-called "Nordics" as a chance- 
genetic combination. 

2. Northeast European. Comprising Poland, Lithuania and 
Esthonia and the Great Russias. Often heavy-set with gray or 
gray-blue eyes. Includes the east-Baltic and other types. 

3. Alpine. The rounder-bodied, rounder-headed, predomi- 
nantly darker peoples of the French mountains, across Switzer- 
land, Austria and to the shore of the Black Sea. 

4. Mediterranean. The peoples on both sides of the Mediter- 
ranean, from Tangier to the Dardanelles, and including the 




Arabian Peninsula. Multiple adaptation to malaria (in the form 
of Mediterranean anemia, Favism, Primaquine-drug sensitivity, 
etc. ) are proving of considerable taxonomic value. 

5. Iranian ( or Irano-Mediterranean ) . The bigger, ruggeder 
and beakier peoples of Asiatic Turkey, Iran and the Soviet Union, 
east of the Caspian. 

6. East African. The long-headed, usually quite linear peo- 
ples of East Africa to the Sudan, lighter, less prognathous and less 
broad-nosed than numbers 8 and 9. 

7. Sudanese. Darker than 6, usually not as extreme in nose 
form, or lip eversion as 8. 

8. Forest Negro. Deeply pigmented, spiral-tuft hair, con- 
siderable prognathism, lip eversion, etc. The holotype of the 
"Negro." The people of West Africa and much of the Congo. 

9. Bantu. A recently expanding group of peoples, usually 
lighter in skin color than 8. 

10. Turkic. The pastoralists and oasis farmers of Central Asia, 
heavy-set and broad-faced. 

11. Tibetan. The taller, more linear, quite nasal peoples of 
Tibet, extending northward to Mongolia. 

12. North Chinese. Tall, often linear, frequently with external 
eyefolds. North China and Manchuria. 

13. Extreme Mongoloid. Siberia, Mongolia to the Kamchatka 
Peninsula. Little facial hair, heavy facial fat-padding, snub noses 
with depressed roots, narrow slitted eyes with marked internal 
eyefolds, the holotype of the "Mongoloids." 

14. Southeast Asiatic. South China to Thailand and Burma 
and the offshore islands. A rapidly expanding population of 
generally-small peoples. 

15. Hindu. Light brown to dark-skinned, often resembling 
Mediterraneans (4), endogamous, widely spread over India. 

16. Dravidian. The rugged-faced, broad-nosed, dark-skinned 
peoples of southern India to Ceylon, possibly related to the Aus- 


Amerindian Groups of Local Races 

17. North American. Primarily taller, more rugged hunting 
peoples of Canada and the United States. More Ai than in Cen- 
tral or South America, especially in the Hudson Bay area. Diego- 
positive individuals often rare as compared with 18 below. 

18. Central American. Shorter, agricultural peoples from the 
American Southwest to Bolivia, almost exclusively O, less than 
20% Diego-positive (Di a ). 

19. South American. Primarily the agricultural peoples of 
Peru and Chile. Apparently a higher incidence of Diego-positive 
individuals and more blood type N. 

20. Fuegian. That is the non-agricultural inhabitants of the 
extreme tip of South America (including the Alacaluf, Ona and 

II. Some Isolated Small Local Races 

21. Lapp. The very small-statured, small-toothed, round- 
headed, almost fragile-appearing fishermen and reindeer herders 
of the tundra and swampy areas of West Russia, Finland, Sweden 
and Norway, largely north of the Arctic circle. 

22. Pacific "Negrito." Small-statured, dark-skinned, frizzly- 
haired, actually a series of local populations ranging from the 
Philippines to the Queensland area of Australia and, taken as a 
whole, a geographical race. Despite their appellation, the Pacific 
Negritos have no necessary connection with the African Geo- 
graphical Race or even 23 below. 

23. African Pygmy. Particularly the pygmies of the Ituri 
Rain Forest, whose small stature merits explanation. Most likely 
the product of isolation and selection early in the expansion of 
African populations. 

24. Eskimo. Geographically restricted to the Asiatic and Amer- 
ican far-north, and broken into extremely small populations, evi- 
dencing many physiological adaptations to extreme cold. Ameri- 
can and Canadian Eskimos differ from both Amerindians and 
most Asiatics in the high incidence of non-tasters, in the near- 
absence of BAIB excretors, and in a low incidence of Di a thereby 
suggesting long-continued isolation and natural selection. 


III. Some Long-Isolated Marginal Local Races 

25. Ainu. The legendary "hairy Ainu" of Yezo in northern 
Japan, apparently the remnant of a once much larger pre- Neo- 
lithic population, and antedating the classical Japanese. While 
hairier, more rugged of face, and markedly different from the 
Japanese, North Chinese and Kamchadals, calling the Ainu 
"primitive whites" turns taxonomy into Procrustes bed. 

26. Murrayian and 27. Carpenterian Australian. Two dis- 
tinct groups of populations, one showing numerous adaptations 
to moderate (32° F.) night cold, and the other more nearly tropi- 
cal in adaptations. Both extremely large-toothed, exceeding the 
Neanderthals in this respect, and with skulls resembling Neander- 
thals in many details. 

28. Bushman and Hottentot (Fig. 25). The aboriginal in- 
habitants of South Africa, including ( 1 ) the desert-dwelling Bush- 
men of the Kalahari and (2) the cattle-owning Hottentot. Less 
melanotic than most Africans, small-toothed and exhibiting ex- 
treme glutal fat storage (steatopygia), the extreme peppercorn, 
spiral- tuft hair and early loss of subcutaneous fat. 

IV* Some Hybrid Populations of Known and Recent Origin 

Although all living local races undoubtedly stem from hybrid 
origins, the extent of admixture is generally unknown and popula- 
tions from the same geographical race are ordinarily involved. 
The following four populations, therefore, are of particular in- 
terest because (1) several different geographical races are in- 
volved and (2) because the groups entering into admixture are 
known. Such hybrid populations of recent and known origin are 
of particular utility in the investigation of genetic linkage in man. 

29. North American Colored ( "American Negro" ) . The so- 
called Negro population of the United States, Bermuda and Can- 
ada. Of west- African and northwest-European origin, the accre- 
tion of European genes due to continuing admixture is partially 
balanced by the social phenomenon of "crossing-over." 

30. South African Colored ("Cape Colored"). The analog- 
ous population of South Africa including Bushman-Hottentot and 
Bantu genes with a variable contribution of European genes and 
increasingly some of East Indian origin. 


31. Ladino. Southern European and southern Amerindian, but 
including also (in the breeding population) Amerindians who 
have adopted the Ladino way of life. As with the preceding two 
populations, there is great local diversity. Depending on local 
attitudes, obvious Ladino groups may acknowledge little or no 
Amerindian, or little or no European in their ancestry. 

32. Neo-Hawaiian. A complex of northwest European and 
southern-European with Polynesian and Chinese/Japanese and 
Filipino. The outstanding laboratory situation for the study of 
human racial hybrids under optimum living conditions. 



A-B-O blood group system, 39-41 
incompatibility, 48 

Adaptations, 62-68 

African geographical race, 124 

African Pygmy local race, 130 

Allele — any one of a series of genes having 
the same locus on homologous chro- 
mosomes, 23-52 

American Colored (hybrid) local race, 7, 
50, 131 

Ainu local race, 131 

Alcoholism, 91 

Allison, A. C, 35, 74, 76-77 

Alpine local race, 127 

American Indians (Amerindians), 9 

Amerindian geographical race, 117-118 

Ashkenazic Jews — primarily the Jews of 
northwestern and eastern Europe, 82, 
87-88, 90, 91 

Asiatic geographical race, 121-122 

Australian geographical race, 120-121 

and taxonomy, 49-51 
Diego, 45-46 
Duffy, 44-45 
MNS-U, 41-42 
Rh, 42-44 
Blumberg, B. S., 35 
Body hair, 123 
Body build, 59, 62 

and natural selection, 59, 62 
Body size, 56-59 
Bone mineral, 27 
Boyd, W. C, 12, 43, 50 
Breeding population — a natural popula- 
tion lacking separate endogamous 
groupings (as contrasted with a de- 
mographic population) and there- 
fore possessing a common gene pool, 
Brown, G. M, 63, 67 
Bushman-Hottentot local race, 131 
Bushmen, 66, 131 


BAIB — /?-aminoisbutyric acid, iv, 32, 130 

Baker, P. T., 56, 62, 64, 66-67 

Balding, 26-27 

Bantu local race, 129 

Barnicot, N. A., 31, 63-64, 67 

Beam, A. G., 33 

Behavior and race, 110-115 

Bergmann's law, 59 

Beutler, E., 88 

Bilharzia, 55 

Biochemistry and race, 31 

Birdsell, J., vii, 96 

Blood groups, 39-52 

A-B-O, 39, 41 

and natural selection, 47-49 

Caloric intake, 55, 57 

Cape-Colored local race, 9 

Carabelli's cusp, 28 

Carbonel, V. M., 28 

Central American local races, 130 

Cepellini, R., 73 

Climate, 53-54 

Cold adaptation, 62, 66 

Congenital ectodermal dysplasia — a de- 
velopmental disorder arising in the 
third trimester of pregnancy, 82 

Continental race — see Geographical races 

Cooley's anemia (Mediterranean anemia 
or Thalassemia major), 71-72 

Coon, C. S., vii, 12 

Coronary artery disease, 91 





Day, C. B., 106 

Dean, R. R A., 29 

Dentition, 28-29 

Dermatoglyphics — the whorls and lines of 

the palms and soles, 30 
Desert adaptation, 59, 63, 67 
Diabetes, 91-92 
Diego blood type (Di), 45-47 
Dipodomys — the Kangaroo rat, 66 
Disease and culture, 77-79 
Dobzhansky, T. H., 13 
Dravidian local race, 129 
Duffy blood type (Fy), 44-45 

East African local race, 129 

Elephantiasis, 55 

Eskimo local race, 130 

"Ethnic group," 8 

European geographical race, 122-123 

Extreme or northern Mongoloid local 

race, 129 
Eyefolds, 29, 54 

Familial dysautonomia — a hereditary 
population-limited disorder charac- 
terized by autonomic dysfunction, 82 

Familial Mediterranean fever — a heredi- 
tary "periodic" disease primarily re- 
stricted to individuals of Eastern 
Mediterranean origin, 86-88 

Famine, 58 

Fat-free weight, 56-57 

Fat storage, 62 

Favism — a hereditary disorder involving 
an allergic-like response to the broad 
bean (Vicia fava), 82, 90-91 

Fibrocystic disease, 66 

Fischer, R., 34 

Fisher, R. A., 43 

Fore, 84-86 

Forest Negro local race, 129 

Fuegian local race, 130 

Gajdusek, D. C, 47, 84-86 
Gene frequency — the frequency (or pro- 
portion) of a given allelic gene in a 

population, as computed from pheno- 
type frequencies, 23-24 
Genetic drift — non-directed changes in 

gene frequency, 94-99 
Genotype — the genetic makeup of an in- 
dividual as contrasted with his phen- 
otype, 36, 101 
Geographical race — a geographically de- 
lineated collection of local races, 13- 
15, 116-126 
Geographical races 
African, 124 
Amerindian, 117-118 
Asiatic, 121-122 
Australian, 120-121 
European, 122-123 
Indian, 122 

Melanesian-Papuan, 119-120 
Micronesian, 119 
Polynesian, 118-119 
Glass, B., 97, 106 
Glutathione, 89 
Griffin, E, 34 
Growth and race, 29 
G. S. H., 89 


Hammel, H. T., 65 
Hair, 26-27 
Haptoglobin, 32-33 
Heat adaptation, 64-66 
Heller, H., 87 

Hemoglobin — the respiratory pigment 
consisting of heme molecules linked 
to the protein globin, 70-93 

abnormal, 70-93 

A, C, E, S, 73-77 

A — the adult type of hemoglobin 

C, D, and E — rare, abnormal hemo- 

S — the abnormal hemoglobin found in 
sickle-cell anemia 
Hindu local race, 129 
Hirsutism, 26 
Hulse, F. S., 105 
Hunt, E. E., Jr., 5 
Hybrid population, 131-132 

Ladino, 132 




Neo-Hawaiian, 132 
North American, Colored, 131 
South African Colored, 131 
Hybridization, 100-109 
Hypercholesterolemia — a hereditary dis- 
ease involving excessive serum cho- 
lesterol levels, 92 
Hyperurecemia, 92 


Indian geographical race, 122 
Intelligence and race, 112-113 
Iranian local race, 129 


Ashkenazic, 87-88, 90-91 
North African, 17 
Sephardic, 17, 86-91 
Yemenite, 17 


Kalahari Desert, 66 

Kalmus, H., 8 

Kangaroo rat, 66 

Keitel, H. G., 68 

Kidd blood group (Jk), 47 

Kukukuku, 84 

Kuru — a hereditary neuromuscular dis- 
ease restricted to Eastern New Guinea, 
82-83, 86 

Ladino hybrid population, 132 
Lapp local race, 130 
Lasker, G. W., 96 
Layrisse, M., 46 
Leprechaunism, 91 
Leprosy, 84 
Leukemia, 91 
Levine, P., 47 
Linnaeus, C. von, 3 
Livingstone, F. B., 77-78 
Local race — a breeding population or 
population isolate, 127-132 

African Pygmy, 130 

Ainu, 131 

Alpine, 127 

Bantu, 129 

Bushman and Hottentot, 6, 131 
Central American, 130 
Dravidian, 129 
East African, 129 
Eskimo, 130 

Extreme Northern Mongoloid, 129 
Forest Negro, 129 
Fuegian, 130 
Hindu, 129 
Iranian, 129 
Lapp, 130 

Mediterranean, 127-129 
Murrayian and Carpenterian Austra- 
lian, 131 
Northeast European, 127 
Northwest European, 127 
North American, 130 
North Chinese, 129 
Pacific Negrito, 130 
Southeast Asiatic, 129 
South American, 130 
Sudanese, 129 
Tibetan, 129 
Turkic, 129 


Malaria, 55, 70-93, 88-89, 91, 93 

Mann, G., 59 

Matson, G. A., 44, 47 

Mayr, E., 13 

Mediterranean anemia — see Thalassemia 

Mediterranean local race, 127-129 

Melanesian-Papuan geographical race, 

Melanin — the complex pigments giving 
rise to the characteristic color of the 
hair and skin, 25-26 

Micronesian geographical race, 119 

Micro-race — a highly localized, statisti- 
cally-distinct isolate in which dis- 
tance is the primary reproductive 
barrier, 13, 18-20 

MNS-U blood group, 41-42 

Montagu, Ashley, 8 

Motulsky, A. G., 86, 91 

Murrayian and Carpenterian Australian 
local race, 13 




Natural selection and race, 53-69 

and blood groups, 47-49 
Neel, J. V., 71-72 
Negro, 53-69, 116-132 
Neo-Hawaiian (hybrid) population, 132 
Newman, R., 57, 62 
Northeast European local race, 127 
Northwest European local race, 127 
North American local race, 130 
North Chinese local race, 129 


Ossification, 29 

Race mixture, 100-109 

Race, R.R., 44, 47 

Races, sympatric — races occupying a com- 
mon territory, 23 

Racial types — chance combinations of 
phenotypic traits, 5-6 

Raper, A. B., 74 

Rensch, B., 13 

Rh (Rhesus) blood group, 43-44 
incompatibility, 48 

Rheumatic heart disease, 91 

Rife, D., 30, 106 

Roberts, D. F., 57-58, 106-107 

Rosen, A. P., 90 

Pacific Negrito local race, 130 

Palm folds, 30 

Pheomelanins — the yellow-red melanins, 

Phenotype frequency — the frequency of a 

given phenotype in a population — 
Pigmentation and natural selection, 55-56 

and race, 25 
"Plastic" traits, 23 
Politzer, W. S., 106 
Polycystic disease, 66 
Polygenic trait — one due to multiple 

genes, 24 
Polymorphism — m u 1 1 i p 1 e phenotypes 

within a population — 
Polynesian geographical race, 118-119 
Population, breeding, 6, 24 
Population isolate, 6 
Primaquine drug sensitivity — hereditary 

sensitivity to Primaquine, acetanilid, 

sulfanilamides, etc., characterized by 

glutathione instability, 88-89, 90-91 
Pygmies, 10, 17, 24, 59 
PTC (phenylthiocarbamide) , 34-35 


Race, geographical — a geographically de- 
lineated collection of local races, 
13-14, 116-126 

Race, local — corresponding to the breed- 
ing population or population isolate, 

Race, micro—, 13, 18-20 

Sanger, R., 44, 47 

Sansone, G., 90 

Saudi Arabia, 68 

Scanlan, J. J., 90 

Schultz, A. H., 29 

Segni, G., 90 

Sephardic Jews — primarily the Jewish 
populations of North Africa, Spain 
and Portugal and those in Europe and 
the United States of Spanish-Portu- 
guese origin and maintaining intra- 
group reproductive barriers, 17, 86-91 

Sewall-Wright effect— genetic drift, 94-99 

Sickle-cell disease — the homozygous dis- 
ease state characterized by reduced 
ability of the red cells to carry 
oxygen, 73-76 

Sickle-cell trait — the heterozygous condi- 
tion, 73-78 

Silfverskiold's disease, 91 

Skeleton, 27-28 

Skin cancer, 56 

Southeast Asiatic local race, 129 

South African Colored (hybrid) popula- 
tion, 131 

South American local race, 130 

Steatopygia — excessive deposition of fat 
over the buttocks, 36 

17-ketosteroids — neutral ketonic steroids 
with the ketone at the C-17 position, 
primarily of testicular and adrenal 
origin, 31-32 

Strain, 8 




Subcutaneous fat, 62 
Sudanese local race, 129 
Sutherland, E. H., Ill 
Sympatric races — races temporarily oc- 
cupying a common territory, 23 


Taste, 33-35 

Taste-blindness, 33-34 

Taxonomy, 20-21, 49-51 

Tay-Sachs disease — amaurotic familial 
idiocy, 91 

Teeth, 28-29 
shovel-shaped, 29-30 

Temperament and race, 110-112 

Temperature, environmental, 54 

Thalassemia major — Mediterranean an- 
emia, 71-73 

Thalassemia minor — the heterozygotic 
state, 71-73 

Third molars, 31 

Tibetan local race, 129 

Tooth size, 28 

Trait frequency — the frequency of a par- 
ticular trait in a population, 23 
Turkic local race, 129 
Types, 5 


Ulcers, 48 

Urine concentration, 66 


Vitamin D, 56 


Washburn, S. L., 8 
Weight, 57-58 
Wiener, A. S., 43 
Wolff son, D., 31 
Wormian bones, 27, 125 
Wright, S., 98 


Yar, 84 
Yaws, 55 
Yemenite Jews, 91 




Date Due 
irned Due 

MAY 26 '62 Ml ^6 7 '1 

'fift 6 




oec«s-< •■ 

M&m 2 1 ' e g 







9 BE 

A1IB_2JLbh4 JUL is m w 

FEB Ofi tm KS "8 



Human races. UGC 
573 G234h 

3 lEbE D11DD D733